Allusion and Allegory: Studies in the >Ciris< 3110446812, 9783110446814

Table of contents :
Contents
Acknowledgements
Introduction
1. The proem
2. Beginning from the beginning
3. The night episode
4. Heroic deeds
5. An epic voyage
6. The metamorphosis
Conclusion
Bibliography
Index of passages cited

Citation preview

Boris Kayachev Allusion and Allegory

Beiträge zur Altertumskunde

Herausgegeben von Michael Erler, Dorothee Gall, Ludwig Koenen und Clemens Zintzen

Band 346

Boris Kayachev

Allusion and Allegory Studies in the Ciris

ISBN 978-3-11-044681-4 e-ISBN (PDF) 978-3-11-044776-7 e-ISBN (EPUB) 978-3-11-044712-5 ISSN 1616-0452 Library of Congress Cataloging-in-Publication Data A CIP catalog record for this book has been applied for at the Library of Congress. Bibliographic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available on the Internet at http://dnb.dnb.de. © 2016 Walter de Gruyter GmbH, Berlin/Boston Printing: Hubert & Co. GmbH & Co. KG, Göttingen ♾ Printed on acid-free paper Printed in Germany www.degruyter.com

To Nikolay Epplée wisest and best of my unofficial teachers

It is easy to forget that the man who writes a good love sonnet needs not only to be enamoured of a woman, but also to be enamoured of the Sonnet. C.S. Lewis

Acknowledgements This book is a slightly revised version of my doctoral dissertation, written at the University of Leeds in 2009‒2012. Robert Maltby, who supervised it, did more than I can acknowledge to mitigate the effects of my stubborn ignorance with his patience and wisdom. Gregory Hutchinson (Christ Church, Oxford) and Steven Green (University College London), who examined the thesis, and Regine May (University of Leeds), who read an early draft of a chapter, gave their comments and suggestions. The Department of Classics supported my research by providing a scholarship as well as a friendly and peaceful environment for undisturbed study. Floris Overduin (Radboud University, Nijmegen) and Gail Trimble (Trinity College, Oxford) kindly shared with me their unpublished theses. My parents, Luiza and Alexander Kayachev, and friends, whom I do not attempt to list, were generous in whatever they could assist me with during those years of my helpless devotion to a forgotten Latin poet. This work would also have been impossible without the combined efforts of all who had taught, advised and inspired me before I came to Leeds. Among a multitude of teachers, Andrei Rossius (Moscow University, at the time), Thomas Poiss (Humboldt University, Berlin) and Nikolay Grintser (Russian State University for the Humanities, Moscow) had been my best guides in the ways of classical scholarship. Finally, Richard Davies (University of Leeds) bravely read and corrected the entire manuscript (all remaining errors are mine, as is non patrii sermonis egestas). To them all, named and unnamed, I owe a debt of gratitude, to each of a different kind, but to everyone of a scope that exceeds my power of expression. Quis loquetur potentias Domini, auditas faciet omnes laudes ejus? Easter 2016, Trondheim

Contents Acknowledgements Introduction

VII

1



The proem (lines 1‒100)



Beginning from the beginning (lines 101‒205)



The night episode (lines 206‒348)



Heroic deeds (lines 349‒385)



An epic voyage (lines 386‒477)



The metamorphosis (lines 478‒541)

Conclusion Bibliography

20

209 212

Index of passages cited

224

92

128 145 173

55

Introduction 1 The Ciris is a notoriously controversial poem, and perhaps it is no exaggeration to say that the scholarly debate about it, the Cirisfrage, is better known than the text itself. Although the issue of the poem’s authorship dates back to the sixteenth century,¹ Skutsch was the first to offer a systematic treatment when, at the very beginning of the twentieth century, he argued that the Ciris is a work by Gallus.² Skutsch’s case instantly provoked a vigorous polemic,³ and was eventually rejected, but only to be revived by Gall at the very end of that century.⁴ The two monographs thus establish a convenient chronological framework (1901 to 1999) in which to situate the Cirisfrage, though, needless to say, it still remains without a decisive answer. The Cirisfrage would on its own make a perfect topic for a major study on the history of classical scholarship, whereas here I can only present a concise critical outline, insofar as it may be relevant to my exploration of the poem itself. In his first book Skutsch developed three principal kinds of argument in favour of the temporal priority of the Ciris over Virgil. First, he interpreted Virgil’s sixth and tenth eclogues as Kataloggedichte paying homage to Gallus’ poetic achievement and pointed out that the Ciris exhibits numerous parallel passages with both Virgilian poems. Second, he observed that the style (broadly understood) of the Ciris points to a pre-Virgilian date of composition.⁵ Third, he argued that a number of parallel passages shared by the Ciris and Virgil would seem more likely to have originated in the former. Leo, who was Skutsch’s most rigor-

 For example, Scaliger ()  defends Virgil’s authorship. The problem never ceased to attract attention, see e. g. Eglinus (), Voss () ‒, Waltz () ‒, Kreunen () ‒, Ganzenmüller () .  Skutsch (). As Skutsch acknowledges, on pages ‒, the attribution to Gallus likewise dates back to the sixteenth century.  To cite only publications directly related to the Cirisfrage: Eskuche (), Helm (a; b), Jahn (), Knaack (), Leo (), Sonntag (), Galdi (), Skutsch (), Vollmer (), Hardie (a; b), Jahn (), Körte (), Leo (), Némethy (), Sudhaus (), Drachmann (), Jahn (), Belling (), Kolář (). Jahn () ‒, , () ‒, ‒, () ‒ offers a useful survey of the course of the discussion over some ten years after Skutsch ().  Gall (). Surprisingly, Gall’s monograph has only received one review, Tschiedel ().  Cf. Duckworth’s () ‒ subsequent analysis of the Ciris’ metre, leading him to the conclusion that there is “every indication that the Ciris is to be dated in the late Republic; it should not be considered post-Vergilian” (p. ).

2

Introduction

ous and, as it turned out, successful opponent, reacted with an article in which he addressed the same three arguments.⁶ He denied the force of the first argument by rejecting the theory of the sixth and tenth eclogues being Kataloggedichte; he conceded the validity of the second argument; and as for the third argument, he argued that, on the contrary, an examination of parallel passages proves Virgil’s priority. The logical conflict between the second and the third points he explained away by suggesting that the Ciris is a product “eines zurückgebliebenen Neoterikers.”⁷ Skutsch replied to the critique by Leo (and others) with a second book in which, though not giving up his first and second arguments, he mainly concentrated on refuting Leo’s demonstration of Virgil’s priority on the basis of a scrutiny of parallel passages.⁸ Another article by Leo followed, offering yet another re-examination of parallels between Virgil and the Ciris. ⁹ We do not know how, and in fact whether at all, Skutsch would have responded (he died in 1912), but thirty years later Helm wrote “Ein Epilog zur Cirisfrage”¹⁰ in which he recognised the comparative analysis of parallel passages as the only method of determining the relative chronology. It only remained to apply the same method to the (much less numerous and precise) parallels between the Ciris and Ovid¹¹ and then those (still less numerous and precise) between the Ciris and Silver Latin poets,¹² which would put the date of composition in the second or even third century AD.¹³ In her monograph Gall returned to parallels between the Ciris and Virgil, but she also reintroduced, though in a modified and more subtle form, Skutsch’s other two arguments, which may be called ‘structural’ and ‘stylistic’. First, she observed that sometimes (most conspicuously in the eighth eclogue) parallels with the Ciris are arranged not sporadically, but in a certain pattern following the internal structure of the Virgilian context, which points to Virgil’s alluding to the Ciris rather than the other way round. Furthermore, instead of arguing that the sixth and tenth eclogues are Kataloggedichte as Skutsch had done, Gall merely noted that parallels with the Ciris are often concentrated in contexts in some way associated with Gallus, not only in Virgil’s eclogues, but also, for  Leo ().  Leo () .  Skutsch ().  Leo ().  The title of Helm ().  Munari ().  Lyne ().  Cf. Lyne () ‒. But note that later, as is attested by Hutchinson () xii, “Lyne came to doubt […] the dating (indeed, metrical and other features recommend an Augustan date).”

1

3

example, in Propertius 1.5, which points specifically to Gallus’ authorship. Second, she demonstrated that in the majority of cases passages shared by the Ciris and Virgil are stylistically more typical of the former than of the latter, which, again, is more consistent with the Ciris being the source. This is obviously not the place to present a thorough re-examination of the Virgilian parallels in the Ciris. It seems appropriate, however, to offer some considerations on why a method of establishing the relative chronology based primarily on a comparative analysis of parallel contexts may be less reliable than one would hope, especially when applied to a case like that of the Ciris and Virgil. The principle is conveniently formulated by Tarrant: “if two passages are verbally similar in a way that precludes coincidental resemblance, and one is organically related to its context while the other is not, the former is the original and the latter the imitation.”¹⁴ Although it may be true that in reality the original often fits in its context more organically than the imitation, I would argue that in principle the reverse inference is false. Abstractly speaking, it is evident that if, for example, one text is poetically better than the other, the better text will be likely to incorporate the shared passage more organically, whatever the chronological relation. If, however, we assume that both texts are equally good, it is probable that whatever faults we find in either of them result from our lack of understanding. It is clear, accordingly, that the Ciris is in a weaker position than Virgil both, so to speak, objectively and subjectively. On the one hand, it can hardly be contested that Virgil is a better poet than the author of the Ciris (even if it is Gallus). On the other, Virgil’s poems are incomparably better understood than the Ciris, for a multitude of reasons, but first of all because they have been given much more scholarly attention. The validity of these two general points can be supported with two more specific arguments. Both Skutsch and Gall emphasise Virgil’s taste for borrowing substantial material (sometimes whole lines) from his predecessors with only small changes, which they suggest refutes the objection that Virgil would not quote extensively from a poem like the Ciris. ¹⁵ I propose to look at this Virgilian practice from a slightly different perspective and consider its conceptual background. It seems clear that this device differs from ordinary allusion, for which extended literal quoting is not required (as well as from ordinary imitatio cum variatione). What then is the point of such a tour de force? Virgil’s anecdotal reply to those who accused him of plagiarism (VSD 46 facilius esse Herculi clauam quam Homero uersum surripere) may suggest a more precise answer than is gen-

 Tarrant () ‒, summing up Axelson ().  Skutsch () ‒ and () ‒, Gall () ‒.

4

Introduction

erally realised. Needless to say, ‘to steal a line from Homer’ can only mean ‘to reuse a Homeric line in such a way that it would fit in the new context as perfectly as the original’. If this is how Virgil conceived of his extensive borrowings, it is obvious per se that he would only quote a passage from the Ciris if he could make it fit the new context better. It has been pointed out that the Virgilian mot is unlikely to be authentic since the accusation of plagiarising Homer would most naturally have been directed at the Aeneid, which was only published posthumously.¹⁶ Whether or not the wording is authentic, the idea sounds genuinely Virgilian, or at least characteristic of Virgil’s intellectual milieu. Virgil’s practice of reusing entire lines from earlier (Latin) poetry with only minimal changes can evidently be related to the widespread critical method of metathesis and perhaps in particular to the view of metathesis taken by Philodemus (and Lucretius). In general, the most straightforward application of the method, already attested in Plato (Phdr. 264c–e), is this: if a poem allows of a transposition of its parts that does not destroy the whole, it is a proof of the poem’s lack of organic unity.¹⁷ Looking back at the Virgilian reply, we can see that the reason why it is especially difficult to steal from Homer is because his poems (are thought to) manifest such a high degree of organic unity that no single line can be as much at home in any new context as in its original place. More specifically, the concept of metathesis has a particular relevance to the Epicurean ‘atomistic’ poetics expounded by Philodemus (and practiced by Lucretius).¹⁸ According to this poetics, the composition of a poem must possess organic unity at all levels, from that of whole episodes to that of individual words and even letters, so that a change of the smallest element would radically alter the entire poem; in fact, it would become a different poem altogether. We can apparently see this principle illustrated by the verbatim repetition, in whatever direction, of a passage of four lines shared by Ciris 538‒541 and Georgics 1.406‒409: although no change occurs within the passage itself, it takes up a different meaning merely by virtue of being put into a new context.¹⁹ If this is the conceptual background behind Virgil’s practice of borrowing whole lines and even groups of lines, the argument from organic unity can obviously not be used for determining the relative chronology in any straightforward manner.

 McGill () .  Cf. De Jonge ()  n. , ‒.  See especially Armstrong ().  Cf. Kyriakidis’ () ingenious suggestion that the repetition of the so-called second proem of Lucretius’ first book (.‒) as the proem of the fourth book (.‒) may likewise be intended as an illustration of that principle.

1

5

While this ‘objective’ factor may on its own be sufficient to make such a method of establishing priority ineffective, it is important to bear in mind that the ‘subjective’ factor, too, seriously affects its applicability. It is difficult to overestimate, and in fact it has proven to be all too easy to neglect, the cumulative impact of our lack of understanding of many aspects of the Ciris in comparison with Virgil’s poems. Here I only propose to consider one such aspect, perhaps the most tangible one: the use of Greek poetic sources. It may seem logical that, if a specific passage shared by the Ciris and Virgil appears to be modelled on a Greek context, that passage probably originated in the text which is more likely to allude to that particular Greek model. The difficulty is, however, that we are much more familiar with Virgil’s use of Greek texts than with that of the Ciris poet, which manifests itself in two ways. First, when we have to decide whether it is Virgil or the Ciris that alludes to a particular Greek context, Virgil’s connection with it is likely to appear more natural just because it is more familiar. Second, when a shared passage has a potential Greek model which better suits the Virgilian context, it may also have an alternative Greek model which would better suit the Ciris context, but of which we are simply unaware. A few examples follow (I discuss them in greater detail in later chapters). Perhaps the best-known case is the parallel between Ciris 280 aut ferro hoc (aperit ferrum quod ueste latebat) and Aen. 6.406 at ramum hunc (aperit ramum qui ueste latebat). Skutsch, following Norden’s suggestion, argued that the Ciris context must be the model since it is more natural to hide in one’s clothes a blade than a Golden Bough.²⁰ In the commentary on Aeneid 6, however, Norden changed his mind as he had discovered that the Bough is modelled on the φάρμακον Medea hides in her breast-band (3.867 κάτθετο μίτρῃ, 1013 ἔξελε μίτρης).²¹ Indeed, Virgil’s systematic use of Apollonius is well-documented; yet Gall is right to point out in response that the Argonautica would likewise be a natural model for the Ciris,²² and as is argued later on, Scylla’s ferrum also has other points of resemblance with Medea’s φάρμακον (see below Chapter 3§6). Similarly, Virgil’s indisputable allusion to Od. 4.406 πικρὸν ἀποπνείουσαι ἁλὸς πολυβενθέος ὀδμήν at Georg. 4.431 exsultans rorem late dispergit amarum has been used as an argument for the priority of the Virgilian context over Ciris 516 et multum late dispersit in aequora rorem. ²³ Yet it can be argued that the Ciris passage finds an alternative model in Call. H. 4.14 Ἰκαρίου πολλὴν ἀπομάσσεται ὕδατος ἄχνην and its earlier reuse at Lucr. 1.719 Ionium glaucis aspargit uirus ab undis (5§4). A third case is the    

Skutsch () . Norden () , originally published in . Gall () ‒. Jahn () ‒. Cf. contra Skutsch () ‒ and Gall () ‒.

6

Introduction

parallel between Ciris 437 omnia uicit amor: quid enim non uinceret ille? and Ecl. 10.69 omnia uincit Amor: et nos cedamus Amori, where the Virgilian context is without a close Greek analogue, whereas the Ciris passage is arguably modelled on Call. H. 1.74‒75 ὧν ὑπὸ χεῖρα […] πάντα· τί δ’ οὐ κρατέοντος ὑπ’ ἰσχύν; (5§9). Finally, a more complex example is provided by ut uidi, ut perii, ut me malus abstulit error, a line that occurs in exactly the same shape at both Ciris 430 and Ecl. 8.41. It is obviously very close to both Theocr. 2.82 χὠς ἴδον, ὣς ἐμάνην, ὥς μοι πυρὶ θυμὸς ἰάφθη and 3.42 ὡς ἴδεν, ὣς ἐμάνη, ὣς ἐς βαθὺν ἅλατ’ ἔρωτα, and naturally it has been argued that a Theocritean allusion is more likely to have originated in the eclogue.²⁴ The Latin line, however, may also have a third Greek model in Mosch. Eur. 74 ὥς μιν φράσαθ’ ὣς ἐόλητο, with perii rendering ἐόλητο (understood as a form of ὄλλυσθαι), which would point at the priority of the Ciris since it is demonstrably influenced by the Europa both in this context and elsewhere (5§7). This brings us to a further complication: poets like Virgil (and the author of the Ciris, for that matter) often simultaneously allude to several models, some of which may be less relevant for the given context than others. If, therefore, a formal argument from comparative analysis of parallels is unlikely to provide a reliable conclusion about the relative chronology of the Ciris and Virgil, can the Cirisfrage be solved in any other way? I suspect that unless we find a new piece of external evidence unambiguously indicating the author, positive proofs of the Ciris either predating or postdating Virgil can hardly be established.²⁵ But this does not mean that a plausible solution is impossible. In fact, I would suggest, the debate about the Ciris has greatly suffered from the desire to find positive proofs, which has often resulted in overstating some arguments and neglecting others. What is rather needed, I believe, is a holistic interpretation of the poem that would attempt to situate it at a particular moment within the poetic tradition. As has already been pointed out, a post-Virgilian dating of the Ciris necessarily involves an intrinsic paradox, namely the conflict between the poem’s (assumed) date of composition and its neoteric style. Obviously any reading of the Ciris as a post-Virgilian poem had to face this paradox, and it seems possible to distinguish, in rough terms, two approaches. One line of interpretation has seen in the poem’s neoteric colouring either a personal idiosyncrasy of the author or an anachronistic fashion of the literary circle he belonged to. It goes  Jahn () ‒. Cf. contra Skutsch () ‒ and Gall () ‒.  Gatti’s () argument that Ciris  alludes to the fact that Octavian was born under the constellation of Libra, which would have been unlikely before the battle of Actium ( BC), seems unconvincing.

1

7

hand in hand with the view of the Ciris as a deeply derivative poem whose neoteric elements are a result, not of creative imitation, but of extensive borrowing by a plagiarist from authentic neoteric poetry.²⁶ This trend culminated in a fundamental commentary by Lyne, for whom “the method of composition of the Ciris was approaching that of a cento.”²⁷ More recently another theory has developed, which interprets the Ciris as a playful neoteric-style forgery pretending to be a juvenile work of Virgil.²⁸ In comparison with the first approach, it has the advantage of viewing the Ciris as sophisticated poetry, though rather sui generis, and of explaining its extensive use of Virgil. This is not the place to discuss these two interpretations in detail, but it seems obvious that in either case the Ciris is only allotted a marginal place within the poetic tradition (whether rightly or not, is another question).²⁹ The above paradox can be avoided by assuming a pre-Virgilian date of composition, which will have the extra advantage of placing the Ciris in a distinct and understandable literary context. Rather than offering another set of straightforward arguments for the priority of the Ciris, this study undertakes what may be described as a thought experiment. I propose to accept a pre-Virgilian dating ex hypothesi and, accordingly, to read the Ciris as if it were written before Virgil. Needless to say, such a reading can itself be an argument either for or against the underlying assumption, depending on how cogent it proves to be. My exploration will mainly focus on pre-Virgilian subtexts of the Ciris, both Latin and Greek, which have received considerably less attention than parallels with Virgil. At every stage, the discussion will essentially proceed in two movements: first noting a subtext and then interpreting it. The two basic concepts underlying this approach, allusion and allegory, are discussed in the next section.

 For the (highly conjectural and, in my view, largely illusory) exploration of the Ciris’ neoteric sources, see Knaack (), Sudhaus (), Ehlers (), Lyne () ‒, Thill () ‒, Thomas (), Knox ().  Lyne () . Though later, to quote Hutchinson () xii again, “Lyne came to doubt […] the excavation of intact neoteric poetry,” too.  Bretzigheimer (), Peirano () ‒, to an extent anticipated by Gorman ().  A number of recent (mostly minor-scale) studies may be mentioned which demonstrate that the Ciris can successfully be interpreted by the same standards as mainstream poetry: e. g. Pigeaud (), Connors (), Bartels (), Woytek (), Faber (), De Gianni ().

8

Introduction

2 Allusion is a modern concept (though it seems to describe the same reality as the more traditional terms ‘imitation’ or ‘borrowing’), and indeed the recent discovery of this device brought about what can perhaps be called a revolution in the understanding of, above all, Hellenistic and Roman poetry, comparable in its impact to the dissemination of the oral-formulaic theory in the field of Homeric scholarship – to this parallel we shall return shortly. There are now numerous studies of individual allusions (or complexes of allusions), of which the two mutually supplementary fundamental works on Virgil’s interaction with Homer and with Apollonius by Knauer and Nelis perhaps deserve to be mentioned more than any others.³⁰ There have also been attempts, within different frameworks, to establish formal principles for the description and classification of allusions, and here Wills’s monograph on the relation of allusion to rhetorical figures as its material base can be singled out from a wide variety of approaches.³¹ Finally, and most relevant to our present concerns, allusion has been treated in a number of book-length studies from a theoretical perspective; three scholars may be named whose combined research can be taken as fairly representative, and in fact formative, of classicists’ present-day theoretical views on allusion: Conte, Hinds, and Edmunds.³² It is their ideas, in a generalised form, that I presuppose and build upon in the following discussion, though often without referring to them explicitly. Students of allusion often ask how allusion is to be interpreted, or sometimes what it is, but very seldom – why it is. For instance, in the case of a Virgilian allusion to Theocritus, scholars will be concerned with questions like how the meaning of that Virgilian context is enriched by this allusion, or perhaps what the connection between these two passages we call allusion actually is, how it comes into being and where it exists (in the text, in the author’s mind, or just in the reader’s imagination), or sometimes why Virgil is here alluding to Theocritus (and not, say, some other poet); but the question why Virgil alludes in the first place is not commonly raised. This does not, however, amount to saying that it is never answered; indeed, the implied answer seems to be that Virgil makes allusions because other contemporary and earlier poets make allusions, too. Yet this answer is only partly true for Virgil taken individually and does not explain why those other poets make allusions. There seems further to exist

 Knauer (), Nelis ().  Wills (). Cf. also Thomas ().  Conte (), Hinds (), Edmunds ().

2

9

among students of Latin and Hellenistic poetry a kind of silent agreement (sometimes stated in passing, but never discussed) that allusion originated in the context of the intimate interaction between poetry and philology that took place in such Hellenistic centres of learning as, first of all, Ptolemaic Alexandria, and that, to a significant degree, allusion is nothing more than the practices of philological commentary and exegesis exercised in poetry. This is a tempting explanation, especially as it renders allusion something intuitively familiar to ourselves as literary scholars; but it is, I believe, basically false. Not only did poets make allusions before the Alexandrian poetae docti, as is documented by the reception of Homer in tragedy and lyric,³³ but there is also some evidence that the Alexandrians themselves were fully aware of the fact and saw in allusion a fundamentally traditional device.³⁴ Let us put this genealogical enquiry aside for a moment, in order to highlight another blind spot. As just pointed out, the fact that other poets knew allusion before Virgil (our test case author) gives only a partial answer to the question we ask, for it only explains where Virgil had learned allusion from, but not why he decided to employ it. To phrase it in more abstract terms, allusion is always studied with the focus on reception and interpretation, but people hardly ever ask what allusion means in respect of composition. For in general the process of composition is something one is reluctant to explore when dealing with poetic masterpieces (and it is mostly masterpieces that we as classicists have and study). Above I suggested that the discovery of allusion had an impact on studies of Hellenistic and Roman poetry comparable to that exercised by the oral-formulaic theory on research on Homer. Now we may return to this parallel and observe that, whereas students of literary ancient poetry mainly focus their attention on issues of reception rather than composition, in the field of early epic studies the situation has largely been the reverse, the oral-formulaic theory being principally intended to explain how the Homeric poems were created rather than how they were (meant to be) received.³⁵ More recently, however, persuasive attempts have been made to develop on the basis of the oral-formulaic theory an aesthetics of reception applicable to the circumstances of oral

 On the reception of Homer in tragedy, see Garner (), Lange (), Kraias (), cf. also the next footnote; in lyric (Pindar), see Nagy (a).  As has been shown by Schmakeit () , Apollonius systematically explores Homeric allusions in Attic tragedy, and moreover, not only does he feature ‘window’ allusions through a tragedian to Homer, but he also follows such ‘window’ allusions created already by Euripides to allude to Homer through Aeschylus.  For this diagnosis, cf. De Jong () ‒.

10

Introduction

performance.³⁶ As a result, our understanding of the oral poetics of Homer paradoxically turns out to be in a sense fuller than that of the literary poetics of, say, Virgil, since oral poetics can now account for both compositional and receptional aspects. Certain insights of the oralists, I believe, can be of relevance to the study of literary poetry as well, and we could profit from having a closer look at what they have to offer us. We may begin by sketching, in rough outline, the mechanics of oral composition. The basic generative principle is composition by theme:³⁷ the epic singer narrates similar situations (‘themes’) using standard sequences of motifs and sets of formulas. The theme is not, however, a simple system of binary choices between either using or not each of the constituent motifs, but rather a highly developed structure comprising multiple variants of every given motif.³⁸ (Although the theme is generally viewed as the oralist counterpart to the traditionalist notion of type-scene,³⁹ the same principle can also be applied to larger narrative units, such as episodes or entire poems.⁴⁰) Thus, the epic singer does not have to produce and memorise in every detail the text of the poem he is going to perform, but only needs a basic outline of the story, whereas the actual wording will be supplied from his repertoire of themes in the process of performance. The same thematic principle, it has been suggested, can explain not only the process of oral composition, but likewise that of reception; in fact, the two mechanisms are arguably very similar.⁴¹ Much as the epic singer, while composing a scene, does not have to deliberate on every single word, but can mostly use standard material supplied by the relevant theme, which gives him time to think about some innovative details he wants to insert into the story, so the listener to an oral performance, likewise familiar with this same theme, does not have to focus his attention on every word he hears, but only on what sounds unexpected. This is, of course, an oversimplification in many aspects, but it adequately illustrates the basic analogy underlying the processes of oral composition and reception. However, some specifications relevant to the ongoing argument can be made.

 See especially Foley () ‒.  Lord () and () ‒, Nagy (b) ‒.  Cf. Nagler () ‒.  So already Lord () : “It is approximately what Arend has called die typischen Scenen,” referring to Arend (). Cf. Edwards () for specifications.  Cf. Nagler () ‒, Louden () xii.  Cf. Kelly’s ()  observation about “the natural symbiosis of composition and reception” in “Archaic hexameter poetry.”

2

11

First of all, the opposition of standard and novel material is to an extent only relative. On the one hand, as already pointed out, the theme is not a rigid and uniform system, so that even an ‘uncontaminated’ realisation of a single theme (if such a thing were possible) would require a certain amount of attention from both the performer and his audience. On the other, novel material, external to a given theme, is a no less integral part of the thematic system, only it comes from a different theme. The latter point may intimate what kind of attention the audience gives to such innovations: an experienced listener will recognise where this or that irregular motif comes from, and thus its original thematic associations will enrich the effect of the narration. This is, basically, a somewhat simplistic formulation of what is called the aesthetics of ‘traditional referentiality’, a central concept in the receptionalist perspective on oral poetry.⁴² Furthermore, the theme is an abstraction that seems useful from the standpoint of a fixed moment, but dissolves into a sequence of numberless performances of similar (and genetically related) scenes when viewed from the perspective of historical development. This consideration has two bearings on our understanding of the referential potential of traditional themes. First, and most importantly, the epic singer (as well as his audience) will remember not only the generic rules of a given theme, but also a certain number of its realisations in individual performances (perhaps the most outstanding, or just the most recent ones), so that he will be able to insert (and his audience to recognise) references not only to the general conventions of a theme, but also to individual details of a particular performance. Second, to return to the dichotomy of standard and novel material, it is now obvious that ultimately, for any traditional motif dictated by a given theme, there was a time when it was an innovation, an intrusion of another theme; therefore, although this motif has become part of tradition, it still preserves, in the genetic memory, its referential potential, even if it is not always realised by a particular performer or audience. Let us recapitulate our (extremely concise) account of oral poetics, by stressing certain fundamental points that are also relevant, as will be argued shortly, to the study of literary poetry. Firstly, the basic existential principle of oral poetry is composition by theme. Secondly, as a consequence, oral poetry is, by virtue of its thematic nature, pervasively intertextual. And thirdly, also as a consequence of its thematic traditionalism, in oral poetry the mechanisms of composition and reception are closely associated and basically isomorphic. All of these three fun-

 See e. g. Foley () ‒, ‒. Cf. further Pucci () ‒, Nagy () ‒, Graziosi and Haubold () ‒, Kelly () ‒.

12

Introduction

damental points, I would argue, can be extrapolated in a meaningful way, mutatis mutandis, to the literary poetic tradition of classical antiquity. This argument will not, however, be as innovative as it may appear at first sight, for it has partly been anticipated by Cairns’s study of the generic, as he calls it, principle of composition in ancient poetry.⁴³ What Cairns calls genres “are not classifications of literature in terms of form” (as genres normally are), but “in terms of content,”⁴⁴ and thus defined they become a concept very similar to that of theme in oralist discourse. As Cairns argues, “the whole of classical poetry is written in accordance with the sets of rules of the various genres,”⁴⁵ and this paradigmatic function ascribed to genres is clearly parallel to the pervasive role of themes in oral composition. Much as the theme of oral poetry can be viewed as a sequence of motifs established by tradition, so Cairns treats the genre as a more or less fixed set of smaller elements which he calls topoi. ⁴⁶ Furthermore, as Cairns demonstrates, these generic rules play a central role not only in composition, but in reception likewise, and, of course, we have just seen that themes, too, underlie both creating and experiencing oral poetry.⁴⁷ Indeed, Cairns himself points out that “from the fifth century BC onwards, the Homeric heroes were frequently regarded as the first exemplars of rhetorical abilities and Homer himself as the inventor of all branches and aspects of the art of rhetoric”

 Cairns (). Cairns has justly been criticised, among other things, for anachronistically taking rhetorical genres as models (in a quite literal sense) for poetic texts, see e. g. Griffin (). Yet Cairns is right to posit that ancient poetry operates with certain structures of expectation that are external to a given text, even if he is wrong in identifying them with rhetorical genres.  Cairns () .  Cairns () .  Cairns () ; to be more accurate, Cairns makes a distinction between primary, constituent, and secondary, optional, elements, of which only the latter he called topoi, but for our present purposes this distinction can be ignored.  See e. g. Cairns () ‒: “The logical incompleteness and apparent internal inconsistencies of many ancient writings are a consequence of their non-individual character, that is, their membership of genres in the sense defined. These writings assume in the reader a knowledge of the circumstances and content of the particular genre to which they belong, and they exploit this knowledge to allow logical connections and distinctions to remain implicit or be omitted altogether.” It seems instructive to compare Cairns’s observation, for both similarities and differences, with an analogous observation about omissions of structurally necessary elements in oral performance made by Nagler () : “When confronted with such an omission the singer’s first reaction will usually be to deny it outright, ‘Of course I sang that part!’ Could not this mean that for the singer and his audience, for whom the total effect of the performance is real and any record of its actual words unimaginably abstract, the missing part is actually there, resonating in their minds […]?”

2

13

(one should bear in mind that Cairns treats his genres in the context of ancient rhetorical theory), and moreover, that “there is plentiful evidence that the Homeric poets were conscious of generic matters, that with full self-awareness they frequently wrote set-pieces, and that sometimes these set-pieces could be sophisticated generic examples,” and finally, that “the early Greek lyric poets showed their recognition of the Homeric generic examples by consciously using and varying the Homeric models of the different genres.”⁴⁸ Why do Cairns’s genres look so similar to Lord’s themes? One answer is obvious, and is already implied in Cairns’s argument: because Homer is the ultimate model of almost all ancient poetry, a model perpetually imitated, both consciously and intuitively. Yet this is only a partial explanation, and there seems to exist a more fundamental reason to account for such a conspicuous continuity of compositional principles – manifested in both oral and written poetry; it may have to do with something in the very nature of poetic creativity. As has recently been demonstrated by Minchin, Homeric type-scenes, or themes, are expressions of what in cognitive theory is called ‘script’, that is the basic and universal mnemonic mechanism responsible for storing in the memory life experience “in terms of standardized, generalized episodes, or structures of expectations.”⁴⁹ And so, according to Minchin, are Cairns’s genres as well: Cairns uses the term ‘genre’ to describe conceptual phenomena which represent “standard responses to standard situations” (34); he claims that the genres which he identifies in ancient literature originate in “important, recurrent, real-life situations” (70). I suggest that Cairns’s genres (universally recognizable verbal routines) may be identified with verbal scripts.⁵⁰

Perhaps the most conspicuous illustration of the continuity between oral and literary compositional techniques can be found in the late-antique Homeric centos by the Empress Eudocia, as analysed by Usher. The key point of Usher’s argument is that “Eudocia, taking her cue from Homer, was actually composing by theme, that is, that she was reusing lines in typical scenes under similar narrative conditions.”⁵¹ I find Usher’s demonstration compelling, but not his explanation. Usher claims that Eudocia “is heir to the ancient tradition of rhapsodic

 Cairns () ‒. Cairns (on p.  n. ) cites Arend () for an evidence of the ‘generic’ character of Homeric poetry (as we may remember Lord did to parallel his notion of theme, see above n. ), but does not even mention Lord, nor any other exponent of the oral-formulaic theory.  Minchin () .  Minchin ()  n. , referring of course to Cairns ().  Usher () .

14

Introduction

performance”⁵² in the most literal way, that is that she actually learnt her compositional skills from rhapsodes. I would rather suggest that Eudocia succeeded in recreating the thematic pattern of traditional oral epic not because she could observe the technique of composition by theme in life (which seems improbable), but because this technique is based on universal principles of the operation of the memory.⁵³ Indeed, as is pointed out by Louden, “although Lord and his followers argue that thematic structure is evidence of orality, we know that certain literate writers tend to organize epics around similar methods and principles.”⁵⁴ The second point was that the thematic principle of organisation results in oral poetry in its pervasive intertextuality; is this likewise true of literary poetry? There are, as we remember, basically two mutually supplementary ways of treating the notion of theme: on the one hand, the theme is a set of abstract rules prescribing how to compose a scene of a certain type; on the other, the theme is a multitude of similar scenes remembered from actual performances.⁵⁵ In every composition the theme can be seen to operate in both ways simultaneously, but their proportional importance may vary depending on the demands of the tradition, the abilities of the singer, and other circumstances. Accordingly, similarities between two analogous contexts resulting from their common thematic nature can be treated as either generic, if the theme has served as a generic model for both these contexts, or individual, if the theme has supplied one of the contexts as an individual model for the other. The same dichotomy can in principle be applied to literary poetry as well.⁵⁶ On the one hand, the genre (in Cairns’s sense) is an ideal entity with certain, often implicit, generative rules; on the other, it consists of numerous concrete exemplars sharing certain characteristic features. Accordingly, between two analo-

 Usher () .  To do Usher justice, it should be acknowledged that he does make a connection between the compositional technique of Eudocia’s centos and certain basic principles underpinning mnemonic activity in general: see Usher () ‒.  Louden () .  Such a view is perfectly compatible with the theme’s ‘scriptic’ nature, for the script is likewise responsible for both (‘intuitively’) remembering generic rules of a standard situation and (‘deliberately’) recalling in one’s mind its individual occasions, cf. Minchin () ‒.  Cf. Kelly () : “Traditional referentiality […] is, instead, generic. This is not to say that more specific referentiality is an invention or characteristic feature exclusively of literate culture […]. There is, therefore, a basic continuity between oral and literate semantics.” In other words, it is clear that oral poetry tends towards ‘generic referentiality’, whereas literary poetry towards ‘specific referentiality’. The point is that both kinds of referentiality are fundamentally interrelated. Cf. also Foley () .

2

15

gous contexts there may exist similarities of either generic or individual nature, of which the former are generally called ‘topoi’ and the latter ‘allusions’. However, unlike Cairns who saw a conflict between the generic (topoi) and the intertextual (allusions) approaches to treating such similarities,⁵⁷ I would contend that, in fact, the perception of generic affinity is the driving force of allusion, too, in the same way as the theme in its abstract manifestation makes the oral poet remember not only the general pattern of a type scene, but also a number of its individual realisations.⁵⁸ This is, I believe, the answer to the question we were considering a few paragraphs earlier, namely why Virgil makes allusions. Virgil makes allusions because qua poet he ‘thinks’ by themes (or scripts) – of which Cairns’s genres are only a specific case – and further because he is aware of that fact and does not content himself with merely following abstract generic rules, which would result in topoi, but always remembers individual realisations of a given theme or motif, which results in allusions. One important practical implication conveyed by such an understanding of allusion is that ‘multiple reference’ (and in particular ‘window reference’), to use Thomas’s terms, is not at all a dernier cri of Alexandrian poetics, “the most sophisticated form of the art [of reference]”,⁵⁹ as it is often held to be, but more or less the default mode of allusion as such. My third proposition concerning oral poetry, which I claimed to be also applicable to literary poetry, was that the processes of composition and reception are isomorphic at some fundamental level, by virtue of their shared thematic nature. As already mentioned, Cairns made a similar point when he observed that the reader of ancient poetry would approach any individual pieces with an a priori “knowledge of the circumstances and content of the particular genre to which they belong”,⁶⁰ that is with the same information that had provided the basis for their composition. It seems further possible to suggest that not only can the process of reception be compared with that of composition, but also, conversely, the process of composition resembles that of reception in certain ways; that, in par-

 Cairns () : “Another reason why generic studies are neglected is that they are not as friendly as they might be to two cardinal concepts of classical studies – development and imitation. […] There is certainly a very great deal of direct imitation in ancient literature and classical scholars rightly lay emphasis on it. But generic studies often produce paradoxical results in this field. Authors displaying the closest similarity in their use of topoi or of generic sophistication are frequently authors in whose case imitation is most unlikely to have been involved.”  Cf. the discussion of topoi vs. allusions in Hinds () ‒.  Thomas () . Cf. McKeown () ‒.  Cairns () ‒.

16

Introduction

ticular, making allusions is very much like finding and interpreting them.⁶¹ The poet composing a poem can be described, I suggest, as an imaginative reader of the story he is converting into a poem. Much as the reader of a poem operates with thematic associations in order to recall a vast number of potentially related passages, among which he will find passages actually alluded to, so the poet ‘reads’ his as yet unwritten story in the context of all the poems he has ever come across, and from a multitude of thematic resonances with these poems, he ‘sanctions’ some by actually incorporating them into his own composition by means of allusion. In other words, the poet does not so much ‘invent’ allusions as ‘find’ them already potentially present in the plot – and either accepts or rejects them. This has been, no doubt, an extremely monolithic and ideal picture of the process of literary composition, while the reality is much more diverse and often less ideal. Although this account was principally intended as a heuristic model that should enable us to see allusion as something intrinsically inherent to poetry, I do believe that some Hellenistic and Roman poets actually conceived of their own creative activity along similar lines. Here is obviously not the place to support this claim with a detailed argument, but in the course of the forthcoming discussion I hope to show that the allusive practice of the Ciris poet can meaningfully be spoken of in such terms. Unlike allusion, allegory is an ancient concept (though cognate with the modern notion of interpretation). As a consequence, despite the fact that the phenomenon described by this term has significant affinities with that referred to as allusion, the two are not often discussed together. Rather than offering a systematic overview of the concept(s) of allegory,⁶² I shall just confine myself to highlighting, in a concise manner, some of these affinities. We may begin, as we did with allusion, by asking why Virgil uses allegory. The standard answer will be that in this way Virgil mirrors the practices of earlier interpreters of Homer.⁶³ In other words, compositional allegory is thought to have originated in interpretative allegory. As with allusion, this answer is largely

 Cf. Kallendorf (), calling attention to “the fact that there are two readers operating in allusion: the critic who notices the allusion and the author who wrote it” (p. ).  It may indeed be questionable whether it is historically correct to subsume different ancient practices of allegory (often described by different terms) under one category. For a recent survey of early Greek practices, see Obbink (). Ford () and Laird () trace the origin of allegory back to the circumstances of oral performance. Murrin, in () and (), offers a perceptive treatment of allegorical poetry from classical antiquity to the Renaissance (and even Romanticism).  Cf. e. g. Murrin () ‒, P. Hardie () ‒, Most () ‒.

3

17

true in practical terms, but errs on a fundamental level in implying that allegory is essentially extraneous to poetry. Although it cannot be denied that many actual modes of interpretative allegory are what we would call inauthentic or anachronistic, there is something in the essence of poetry that invites such an approach.⁶⁴ It seems possible to suggest that, like allusion, this implicit allegorical potential can be seen, at least from a certain perspective, as a function of poetry’s thematic nature. In order for a particular literal narrative to allude to a particular allegorical narrative, these two narratives need to possess a generic similarity, that is they have to participate in the same theme. (Just as with allusion, in a given case it will depend first on the poet and then on the reader which of the available allegorical possibilities are actualised.) To put it in more abstract terms, both allusion and allegory are devices that link an actual text to an external, and often implicit, point of reference: in the former case, it is another text; in the latter, it is another, let us say, narrative. This structural analogy of the two devices means that in reality they often coincide when the allegorical narrative referred to happens, so to speak, to be embodied in a particular text. (Needless to say, much as allusion does not as a rule refer to a whole text, but just to a specific context, so any figure or image may not normally reveal the entire allegorical narrative, but only a fragment.) As a consequence, in certain kinds of poetry allusion and allegory can be mutually dependent on each other: allusion providing the material form for allegory, allegory creating the conceptual content for allusion. The Ciris is demonstrably a poem of this sort, and accordingly such a dialectic of allusion and allegory – understood along the lines sketched above – will establish the methodological foundation of the following study.

3 To recapitulate, this study will mainly focus on the Ciris’ engagement with (preVirgilian) literary sources. The view of allusion presented above entails, among other things, that subtexts should be treated not only as independent entities, but also as part of a poetic tradition which constitutes, at least from the poet’s perspective, an organic whole. It seemed natural therefore to aim at creating a  Laird () develops a penetrating argument for viewing “allegory – at least in the limited sense of conveying x by means of y – as being something which is built into rather than read into epic,” which can be applied “not only to Latin epic (as might be expected from some recent studies) but also to Homer” (p. ).

18

Introduction

broader, even if inevitably fragmentary, picture of this tradition rather than to concentrate on a narrow set of primary models. In practical terms this means that our attention will be directed not only at the systematic use in the Ciris of, say, Lucretius and Apollonius (not to mention Catullus), but also at occasional allusions to Homer or Empedocles, for example, as well as at possible parallels with such unexpected texts as Nicander’s iological poems or the fragments of Choerilus of Samos. Although, understandably, some of the proposed connections can only be posited tentatively, the cumulative weight of the collected material is, I believe, sufficient to demonstrate a degree of erudition and sophistication in the use of literary sources not unworthy of what can be expected from a poet like Gallus. Similarly, and perhaps to an even greater extent, interpreting these allusions is often bound to be speculative, and as a rule my interpretations are offered merely exempli gratia (even if this may not be stated explicitly). Yet I would contend that, although specific formulations may not always be exactly correct, the Ciris does use its subtexts in such a way as to create an implicit, allegorical, pattern with a certain poetological message. As is argued throughout this study, the Ciris’ poetological concerns are expressed in two dimensions, one static and one dynamic, so to speak. Within the former, the Ciris can be seen to explore the tension between a materialist (‘Epicurean’ / ‘Lucretian’) and a metaphysical (‘Platonic’ / ‘Empedoclean’) poetics. Within the latter, the poem’s protagonist, Scylla, can be read as a poet figure, an embodiment of the metaphor of the poet as lover. Such poetological concerns, it is further argued, would have been particularly natural in the intellectual milieu of the Late Republic, in the context of the poetic theories (and practices) of Lucretius and Philodemus. The book takes the form of a discursive commentary, with its six chapters discussing the text of the Ciris in a consecutive manner.⁶⁵ In treating individual sections of the poem, however, each chapter, instead of providing a line-by-line

 It may be interesting to point out that this division of the Ciris roughly corresponds to the structure of a five-part tragedy: the proem (lines ‒) plays the role of a hypothesis, as it were, the exposition (‒) performs the function of a prologue (with the apostrophe of ‒ functioning as a parodos?), three episodes (‒, ‒, ‒) produce a gradual increase of pathos (note that the nurse is a typically tragic character), and the final episode (‒), like an exodos, effects a conclusion through the intervention of a deus ex machina. This may imply that the Ciris’ structure is modelled on a scenic version of the Scylla story, cf. Ov. Tr. ..‒ impia nec tragicos tetigisset Scylla cothurnos, | ni patrium crinem desecuisset amor (cf. Ciris ‒). It is further suggestive that two early-modern dramatic adaptations of the Scylla story, largely based on the Ciris, have a five-part structure: a tragedy Scilla by Cesare de’ Cesari (Venice, ) and a musical tragedy Scylla by Theobaldo di Gatti (Paris, ); see Gatti () ‒.

3

19

discussion, addresses the relevant section from a particular thematic point of view. Chapter 1, on the proem (lines 1‒100), considers how the Ciris establishes its place within poetic tradition and introduces its implicit poetological perspective. Chapter 2 (101‒205) follows the creation of the mise en scène and elucidates the conceptual base of the notion of love in the poem. Chapters 3 (206‒348) and 4 (349‒385) deal with the two central episodes that can be said to explore, in a sophisticated and polemical way, the nature of epic action and narration. Chapter 5 (386‒477) views the account of Scylla’s sea voyage from the standpoint of the metaphor of the ship of song. Chapter 6 (478‒541) considers the poetological implications of Scylla’s metamorphosis into the ciris.

1 The proem (lines 1‒100) 1.1 Unlike many better-known ancient narrative poems that more or less instantly plunge in medias res, the Ciris opens with a lengthy programmatic proem. The first half (1‒53), which will receive most of our attention in this chapter, defines the genre of the forthcoming poem, by means of an elaborate recusatio; the second (54‒100) discusses its subject. It is tempting to read the recusatio against the background of the classic Augustan examples of that rhetorical device,¹ but such an approach may obscure rather than illuminate the specific agenda of the Ciris proem. Moreover, if a pre-Virgilian dating is correct, it is the Augustan recusationes that should be interpreted through the Ciris, not the other way round. In that case, where does the recusatio of the Ciris come from, if not ex nihilo? The most crucial moment in the development of recusatio has been seen in the allusion at the beginning of Virgil’s sixth eclogue – “the first formal, fullydeveloped recusatio in Latin poetry”² – to Callimachus’ Aetia prologue, and it is on the character of this link that the interpretation of the process has largely depended. One particularly influential approach has been to view the Aetia prologue as the chief, or even sole, model of Virgil’s recusatio, not only in terms of poetic imagery, but most importantly for its formal aspects, which would make it possible to reduce the entire problem of the genesis of recusatio to a single act of allusion to Callimachus.³ This seemingly solid position, however, has been challenged by Cameron, who argued that “though undoubtedly the principal source of the first Augustan recusatio, in its original form the Aetia prologue is not itself a recusatio at all.”⁴ In our discussion of the Ciris proem, therefore, we should be prepared to look for subtexts that do not exhibit all the formal features of a developed recusatio.

   

As Guenzi () has done. Ross () . See Wimmel (), Clausen (), Thomas (). Cameron () .

1.2

21

1.2 We may begin with the very first passage of the Ciris, which, as has been noticed by many scholars,⁵ contains a manifest Catullan allusion (1‒11):⁶ etsi me, uario iactatum laudis amore irritaque expertum fallacis praemia uulgi, Cecropius suauis expirans hortulus auras florentis uiridi Sophiae complectitur umbra, mensque, ut quiret eo dignum sibi quaerere carmen, longe aliud studium inque alios accincta labores, altius ad magni suspexit sidera mundi et placitum paucis ausa est ascendere collem: non tamen absistam coeptum detexere munus, in quo iure meas utinam requiescere musas et leuiter blandum liceat deponere amorem.

The subtext alluded to is, of course, Catullus 65, a poem that introduces the translation of Callimachus’ Coma Berenices (1‒4, 15‒16): etsi me assiduo defectum cura dolore seuocat a doctis, Hortale, uirginibus, nec potis est dulcis Musarum expromere fetus mens animi, tantis fluctuat ipsa malis… … sed tamen in tantis maeroribus, Hortale, mitto haec expressa tibi carmina Battiadae.

The connection between the two contexts is unmistakable,⁷ but the structural analogy feels disharmoniously inexact. Catullus’ apologetic message is fairly straightforward: although (etsi), because of his personal circumstances, he cannot offer his addressee original poetry, he still (tamen) will send him a translation from Callimachus, with the implication that the latter is inferior to the former. The thought expressed in the Ciris sounds both similar and different: although (etsi), in his present situation, the poet is planning a more serious poem, he still (tamen) decides to finish an earlier and less ambitious project. We see both poets turn to the second-best kind of poetry, but the two formally analogous concessive clauses convey ideas that are almost the opposites of  The most detailed are the discussions by Guenzi () ‒, Woytek () ‒, and Peirano () ‒; cf. also Bellinger () and Lyne () .  Quotations from the Ciris follow Goold’s text in Fairclough and Goold ().  Perhaps it is worth calling attention to the intertextual pun on Hortalus’ name with hortulus.

22

1 The proem (lines 1‒100)

each other: Catullus speaks of the circumstance because of which, the Ciris poet of that in spite of which, they do what they do. One may be tempted to draw the conclusion that Catullus 65 has only provided a model mechanically imitated in the Ciris for no particular reason. But this would be a mistaken inference, for, despite the apparent incongruity, the Catullan allusion is in fact programmatically important. First of all, we may note that Ciris 7 altius ad magni suspexit sidera mundi is a reminiscence of the opening line of Catullus 66: omnia qui magni dispexit lumina mundi. If we follow the hint, we can see that the Ciris, consisting of a clearly demarcated proem and a narrative part, reproduces the pattern of Catullus 65 and 66 taken together.⁸ Like the story of Berenice’s lock of hair, the story told in the Ciris has an aetiological dimension, and moreover the final metamorphosis, explaining the origin of the perpetual pursuit of the ciris by the sea eagle, is explicitly likened to a catasterism. There is in the Ciris, too, a lock of hair of a comparable symbolic significance, which likewise belongs to royalty and is eventually cut off. Furthermore, it may be relevant that the Coma Berenices, of which Catullus 66 is a translation, was included in the Aetia. Catullus 65 seems to allude to Callimachus’ treatment of the Acontius and Cydippe story, which likewise was part of the Aetia ⁹ (we shall have occasion to see how this subtext resonates in the Ciris in 2§11). In fact, Catullus 65 has also been suggested to echo the Aetia prologue,¹⁰ so that poems 65 and 66 taken together (or indeed the whole of 65‒116?) could be viewed as a Latin version of the Aetia in miniature. The available points of contact are admittedly faint, but perhaps not entirely negligible. It has been argued¹¹ that the first word of Catullus 116 (1‒2 saepe tibi studioso animo uerba ante requirens | carmina uti possem mittere Battiadae: cf. 65.15‒16 quoted above), a poem that closes the elegiac sequence opened by Catullus 65, is a rendering of what is believed¹² to be the first word of the Aetia (fr. 1.1‒ 2 πολλάκ]ι μοι Τελχῖνες ἐπιτρύζουσιν ἀοιδῇ, | νήιδες οἳ Μούσης οὐκ ἐγένοντο φίλοι): πολλάκι may likewise be evoked by assiduo in Catullus 65. The second lines have further correspondences: Catullus’ doctae uirgines can be contrasted with the νήιδες Telchines, who are no friends of the Muse (or of the Muses? Hop-

 It is actually far from impossible that  and  are originally one poem. The archetype of the Catullan collection is notoriously inconsistent in indicating divisions between poems, and in particular it did not divide  and , see Butrica () .  See e. g. Hunter (a), Skinner () ‒, Höschele () ‒.  King () ‒.  Barchiesi () .  See Pontani ().

1.2

23

kinson’s Μούσῃς is certainly tempting¹³). The metaphor of the Muses’ offspring (Musarum expromere fetus) may capture Callimachus’ stock-raising imagery (fr. 1.23‒24 τὸ μὲν θύος ὅττι πάχιστον | θρέψαι, τὴ]ν Μοῦσαν δ’ ὠγαθὲ λεπταλέην). Finally, the (conjectural) mention of nightingales by Callimachus (fr. 1.16 ἀη[δονίδες] δ’ ὧδε μελιχρότεραι) can be linked with the nightingale simile in Catullus (65.13‒14 qualia sub densis ramorum concinit umbris | Daulias, absumpti fata gemens Ityli). The last motif may, in fact, create further significant resonances in Callimachus’ poetry. As has been pointed out by Hunter, the Catullan simile finds a parallel in Callimachus’ hymn to Athena where the lament of Tiresias’ mother Chariclo is compared with the nightingales’ song (94 γοερᾶν οἶτον ἀηδονίδων).¹⁴ Yet this is far from the only parallel between Catullus 65 and the Tiresias section of the hymn (57‒136). As Hunter observes, “the style and content of Chariclo’s lament strongly suggest funeral lament. The loss of sight is like the loss of life.”¹⁵ If Tiresias’ blindness can indeed be interpreted metaphorically as death, and Hunter supplies further indications that it should, not only does Catullus’ lament for his deceased brother parallel Chariclo’s lament for her son in that they both are like a nightingale’s song, but also Tiresias’ fate can be compared with that of Catullus’ brother, who has similarly lost the possibility – not only to see, but to be seen, too (82 παιδὸς δ’ ὄμματα νὺξ ἔλαβεν; 89 οὐκ ἀέλιον πάλιν ὄψεαι; cf. 65.8‒9 tellus … nostris obterit ex oculis, 10‒11 numquam ego te … aspiciam posthac?). The same textual parallels suggest another correspondence, that between Tiresias and Catullus – who is bereft of the sight of his brother. This link is further elaborated by the similarity of their reactions to the misfortunes they are afflicted by: the literal stupor with which Tiresias responds to Athena’s anger (83‒84 ἐστάθη δ’ ἄφθογγος, ἐκόλλασαν γὰρ ἀνῖαι | γώνατα, καὶ φωνὰν ἔσχεν ἀμηχανία) provides a counterpart to Catullus’ creativity crisis caused by the death of his brother (65.3‒4 nec potis est dulcis Musarum expromere fetus | mens animi, tantis fluctuat ipsa malis). There is also a possible point of resemblance between Catullus and Chariclo: much as the former is excluded from the company of the Muses figuratively (65.2 seuocat a doctis … uirginibus), the latter will not come near their mountain physically (90 ὦ Ἑλικὼν οὐκέτι μοι παριτέ).¹⁶ This brings us to the fact, recently demonstrated by a num-

 Cf. Harder () ‒.  Hunter () ‒.  Hunter () .  As I have argued in Kayachev () ‒, the Orpheus episode at the end of Virgil’s Georgics is likewise indebted to the Tiresias section of Callimachus’ hymn. Needless to say, the Virgilian nightingale simile (Georg. .‒) has obvious points of contact with the Catullan

24

1 The proem (lines 1‒100)

ber of scholars, that Callimachus’ treatment of the Tiresias story is a variation on the topos of poetic initiation, with transparent Hesiodic references.¹⁷ In a sense, there are two initiations in Callimachus: Tiresias’ as a prophet and Chariclo’s as a mourner; Catullus, as we have seen, parallels both of them by beginning to write elegy (instead of polymetrics and hexameters). To make a full circle, we may point out that the hymn to Athena also features a number of significant parallels with Callimachus’ account of his own initiation in the Aetia prologue,¹⁸ so that apparently Catullus’ (arguable) use of both these texts in poem 65 only reveals their connection. If the above reasons are valid, Catullus 65, to which the opening of the Ciris so clearly alludes, proves to be not a marginal text with merely autobiographical concerns, but a poem directly related to the core of the tradition of learned poetry. The next step is, naturally, to consider whether the Callimachean subtexts of Catullus 65 are relevant to the Ciris. It must be admitted that the Ciris proem features no close textual parallels with either of them, but it seems also clear that its poetological imagery has more in common with Callimachus than with Catullus. The central point of analogy is that both the Ciris and Callimachus speak of a mountain of (poetic) inspiration. In Callimachus – both in the hymn and in the Aetia prologue – it is, of course, Helicon, the mountain of the Muses (though in the hymn their role is taken over by Athena). In the Ciris the narrator appears to climb a different sort of mountain, the mountain of Wisdom, but he does so with the intention to compose a poem, if a philosophical one (eo dignum sibi quaerere carmen). The association of poetry and philosophy that we observe in both the Ciris and Callimachus – for Athena is, of course, the embodiment of wisdom – plays a crucial part in the poetic programmes of both authors. As I have argued elsewhere, in his treatments of the topos of poetic initiation Callimachus assimilates the figure of the poet to that of the philosopher.¹⁹ Even though on the surface the Ciris seems to oppose poetry and philosophy (or rather mythological poetry and philosophical poetry), we shall see soon that this opposition is not meant to be taken at face value.

one. What we have in the Orpheus episode can thus qualify as a ‘window’ allusion through Catullus  to Callimachus’ hymn to Athena.  The most helpful discussions are Heath () ‒ and Ambühl () ‒; see also Hadjittofi () .  Heath () ‒, Heyworth () ‒, Ambühl () , ‒, Kayachev () ‒.  Kayachev () ‒. Cf. also Acosta-Hughes and Stephens () ‒.

1.3

25

1.3 The allusion to Catullus 65 also resonates on another level, as it introduces a whole set of reminiscences from Catullus’ other elegiac poems thematically related to poem 65, namely poems 68²⁰ and 101. Let us begin by observing textual parallels, which we now find mostly concentrated in a later passage of the Ciris (42‒46): sed quoniam ad tantas nunc primum nascimur artes, nunc primum teneros firmamus robore neruos, haec tamen interea, quae possumus, in quibus aeui prima rudimenta et iuuenes exegimus annos, accipe dona meo multum uigilata labore…

To proceed ὕστερον πρότερον Ὁμηρικῶς, we may first consider poem 101, addressed to Catullus’ dead brother (7‒10): nunc tamen interea haec, prisco quae more parentum tradita sunt tristi munere ad inferias, accipe fraterno multum manantia fletu, atque in perpetuum, frater, aue atque uale.

The epigram comes as a (partial) fulfilment of the promise made in poem 65 (12 semper maesta tua carmina morte canam). In poem 68, as an excuse and explanation for not granting his addressee’s request, Catullus offers an account of the grief caused by the death of his brother (11‒14): sed tibi ne mea sint ignota incommoda, Manli, neu me odisse putes hospitis officium, accipe quis merser fortunae fluctibus ipse, ne amplius a misero dona beata petas.

What immediately strikes us here is that both Catullan contexts alluded to by the Ciris imply the death of Catullus’ brother as the raison d’être of the poetry they refer to: in poem 68, the sad story of Catullus’ misery; in 101, the epigram itself. The same reason was also given in Catullus 65 for translating Callimachus rather than writing original poetry: as we have seen, this crucial event of Catullus’ bi-

 It is to an extent irrelevant whether Catullus  is a single poem or a complex of related poems; here I shall primarily be concerned with what is generally singled out as a (i. e. .‒).

26

1 The proem (lines 1‒100)

ography was there presented as parallel to Tiresias’ initiation to prophecy. Catullus’ experience of the death of his brother can thus be interpreted as an initiation into elegiac poetry (i. e. poetry written in elegiac couplets), with maesta carmina (65.12) alluding to the etymology of ἐλεγεῖον from ἐὴ λέγειν or ἔλεος. Although in adapting Catullus 65 the Ciris poet may appear to ignore the death of Catullus’ brother as a mere autobiographical detail, the allusions to poems 68 and 101, centred on precisely that motif, suggest that he did see in it some programmatic importance. Callimachus’ hymn to Athena was one subtext that allowed us to interpret in terms of poetics what otherwise might seem a purely personal fact of Catullus’ biography. Another important model for Catullus’ self-presentation, I shall argue, can be found in two epigrams by Philodemus. Let us have a closer look at Catullus’ account of what happened to him when his brother died, which directly follows the last quoted passage (68.15‒20):²¹ tempore quo primum uestis mihi tradita pura est, iucundum cum aetas florida uer ageret, multa satis lusi: non est dea nescia nostri, quae dulcem curis miscet amaritiem. sed totum hoc studium luctu fraterna mihi mors abstulit…

As I have suggested elsewhere, Catullus’ autobiographical account is modelled on a programmatic epigram by Philodemus (AP 5.112):²² ἠράσθην. τίς δ᾽ οὐχί; κεκώμακα. τίς δ’ ἀμύητος κώμων; ἀλλ’ ἐμάνην ἐκ τίνος; οὐχὶ θεοῦ; ἐρρίφθω, πολιὴ γὰρ ἐπείγεται ἀντὶ μελαίνης θρὶξ ἤδη, συνετῆς ἄγγελος ἡλικίης. καὶ παίζειν ὅτε καιρός, ἐπαίξαμεν· ἡνίκα καὶ νῦν οὐκέτι, λωιτέρης φροντίδος ἁψόμεθα.

 Note that . non est dea nescia nostri (with  dederit duplex Amathusia curam) is the model of Ciris ‒ non est Amathusia nostri | tam rudis.  Kayachev () . For other possible instances of Philodemus’ influence on Catullus, see e. g. Tait () ‒, Landolfi (), Sider ()  s.v. Catullus, Tilg () , Cairns () ‒. Sider () ‒ even speculates that they may have been personally acquainted. Philodemus’ presence in Catullus  is further supported by the allusion in .‒ sed furtiua dedit media [Landor: mira codd.] munuscula nocte | ipsius ex ipso dempta uiri gremio to AP ..‒ καὶ νυκτὸς μεσάτης τὸν ἐμὸν κλέψασα σύνευνον | ἦλθον.

1.3

27

The narrative situation of the two contexts is very similar: both poets present themselves at a crucial moment when they are about to abandon (ἐρρίφθω, abstulit) their juvenile lifestyle and engage in more serious concerns (λωιτέρης φροντίδος, contrast with totum hoc studium).²³ For both, the centre of their earlier lives was love (ἠράσθην and οὐχὶ θεοῦ; cf. non est dea nescia nostri), which admittedly is not surprising at all; what calls for attention is that both use the ambiguous term ‘to play’ to describe this erotic activity, and perhaps more. Both παίζειν and ludere can refer to poetic composition in lighter genres, and this additional connotation seems to be implied in both Philodemus and Catullus. Even if 68.10 munera et Musarum et Veneris may not be literally a hendiadys,²⁴ the two kinds of ‘gifts’ are no doubt related, if only for the reason that Catullus is unable to provide either of them. Poem 65 suggests that here, too, Catullus’ inability to write poetry may be connected with his brother’s death, which is named as the reason for his loss of interest in the pleasures of love. As for Philodemus, a poetological interpretation of παίζειν is invited by a related epigram in which the analogy of life and poetry is conveyed more explicitly through the metaphor of the book of life (AP 11.41): ἑπτὰ τριηκόντεσσιν ἐπέρχονται λυκάβαντες, ἤδη μοι βιότου σχιζόμεναι σελίδες· ἤδη καὶ λευκαί με κατασπείρουσιν ἔθειραι, Ξανθίππη, συνετῆς ἄγγελοι ἡλικίης, ἀλλ’ ἔτι μοι ψαλμός τε λάλος κῶμοί τε μέλονται καὶ πῦρ ἀπλήστῳ τύφετ’ ἐνὶ κραδίῃ· αὐτὴν ἀλλὰ τάχιστα κορωνίδα γράψατε, Μοῦσαι, ταύτης ἡμετέρης, δεσπότιδες, μανίης.

As I argue elsewhere, these two Philodemean epigrams are a seminal model for the topos of conversion from poetry to philosophy that was very popular among Late Republican and Augustan poets.²⁵ This is not the place to repeat the argument in detail, but it is important to recall Sider’s reading of that pair of epigrams as a programmatic statement that subverts rather than asserts a straightforward opposition of poetry and philosophy.²⁶ This has two implications for the meaning of the topos: first, those who use it do not actually intend to abandon poetry in the future; second, they may already at present not be as uninterested (and incompetent) in philosophy as they claim. What Catullus is doing when he

   

Cf. further .‒ cuius ego interitu tota de mente fugaui | haec studia… See Skinner () ‒. Kayachev (). Sider () and ().

28

1 The proem (lines 1‒100)

alludes to Philodemus is attempting to reshape this topos in specifically Roman terms, as an opposition of poetry and, not philosophy, but ‘real life’, of what is called otium and negotium (though it should be noted that Catullus’ negotium consists in his family concerns, exemplified by the death of his brother, rather than in political pursuits, as, for example, in Cicero’s case). But his point is, mutatis mutandis, the same as Philodemus’: a writer of light poetry may in reality be a seriously-minded person (cf. 16.5‒6 nam castum esse decet pium poetam | ipsum, uersiculos nihil necesse est). Just as Philodemus does not stop composing epigrams at the age of thirty-seven, so Catullus does not abandon poetry after the death of his brother. Turning back now to the Ciris, we can see that this opposition is there again rephrased in the original Philodemean terms, as an opposition of poetry and philosophy. An even closer comparandum can be found in Catalepton 5, a poem often ascribed to the young Virgil, which therefore may predate the Ciris (5.5, 11‒14):²⁷ ite hinc, inane cymbalon iuuentutis … ite hinc, Camenae, uos quoque ite iam sane, dulces Camenae (nam fatebimur uerum, dulces fuistis), et tamen meas chartas reuisitote, sed pudenter et raro.

The affinity of the thought expressed in this little poem (again, a variation on the topos of conversion to philosophy) to that of Philodemus’ programmatic epigrams has been noticed in earlier scholarship.²⁸ We can see that the Ciris features parallels with both the Virgilian (?) poem and Philodemus’ epigrams. On the one hand, just as in Catalepton 5, the narrator claims to have abandoned the vanity of his early years (2 irrita … fallacis praemia uulgi, cf. 5.5 ite hinc, inane cymbalon iuuentutis) in order to become an Epicurean (3 Cecropius … hortulus, cf. 5.11 dicta Sironis), and yet he does not entirely give up writing (light) poetry (9 non tamen absistam coeptum detexere munus, cf. 5.13‒14 et tamen meas chartas | reuisitote). On the other hand, we find in the Ciris further Philodemean concepts: longe aliud studium (6), referring to the Epicurean philosophy, corresponds with λωιτέρης φροντίδος; interdum ludere nobis (19) – an occupation suitable for youth, but to be admitted occasionally in mature years – with καὶ παίζειν ὅτε καιρός, ἐπαίξαμεν. Finally, the Ciris is conceived of as a sort of

 Lyne ()  and  notes some points of contact.  See e. g. Sider ()  n. . Cf. further Kayachev () ‒.

1.4

29

epilogue to the light poetry its author composed in his youth (10‒11 in quo iure meas utinam requiescere musas | et leuiter blandum liceat deponere amorem), and in a similar way Philodemus’ epigrams on the relationship with Xanthippe are ostensibly presented as a closure to his poetic career (αὐτὴν ἀλλὰ τάχιστα κορωνίδα γράψατε, Μοῦσαι, | ταύτης ἡμετέρης, δεσπότιδες, μανίης).²⁹

1.4 There is also a third dimension opened by the allusion to Catullus 65. As is pointed out by Guenzi, both Catullus 65 and its imitation in the Ciris feature remarkable similarities with the opening of the Rhetoric to Herennius (1.1):³⁰ etsi negotiis familiaribus inpediti uix satis otium studio suppeditare possumus et id ipsum quod datur otii libentius in philosophia consumere consueuimus, tamen tua nos, C. Herenni, uoluntas commouit ut de ratione dicendi conscriberemus, ne aut tua causa noluisse aut fugisse nos laborem putares.

Both Catullus and the author of the Rhetoric claim to be hindered by certain family concerns (65.1‒2 etsi me assiduo defectum cura dolore | seuocat, cf. etsi negotiis familiaribus inpediti), yet (tamen) they decide to send their addressees a literary work, so that the latter would not be offended by their neglect (65.17‒18 ne tua dicta uagis nequiquam credita uentis | effluxisse meo forte putes animo, cf. ne aut tua causa noluisse aut fugisse nos laborem putares). Whether or not Catullus is alluding to that particular context, there may be a point in his opening poem 65 in this rhetorical manner. As Lyne points out, etsi “is frequently used as an opening word of a speech or letter by Cicero and others.”³¹ Likewise, as Harder comments on the first word of Callimachus’ Aetia, “[t]he use of this adverb here suggests that the prologue was a kind of ‘speech’, because πολλάκις and similar words are often found at the beginning of speeches in Greek drama, Thucydides, and rhetors.”³² If, as I have argued, poem 65 can be viewed as a version of the

 In Kayachev ()  I suggest that the pun on includere ludo at Hor. Ep. .., a context demonstrably using the same Philodemean topos of progress from poetry to philosophy, may be a pointer to Ciris ‒ ludere nobis | et gracilem molli liceat pede cl(a)udere uersum, thus indirectly supporting Philodemus’ presence there, too. As has recently been argued by Höschele () , Hor. Carm. . is likewise alluding to Philodemus’ epigram AP ., and now we may add that its final line features a similar pun (..): cludere lustrum.  Guenzi () ‒.  Lyne () . Cf. Woodman () ‒.  Harder () .

30

1 The proem (lines 1‒100)

Aetia prologue, Catullus may have intended to imitate in that way this stylistic feature of his Callimachean model. Turning now to the Ciris, we can observe that its points of contact with the Rhetoric are by no means confined to what is derived from Catullus 65. Most significantly, it reproduces the tripartite system of oppositions, involving ‘real life’, philosophy and literature, instead of the bipartite one that we could find in Catullus (as well as in Philodemus). As is made clear in the Rhetoric, the main division is that between negotium (what I call ‘real life’) and otium, which in its turn is divided between philosophy and literature. Guenzi seems right therefore in taking the first two lines of the Ciris (uario iactatum laudis amore | irritaque expertum fallacis praemia uulgi) as a reference to political rather than literary activity:³³ the Ciris poet has made the basic choice between negotium and otium, and now has further to decide between philosophy and poetry. It is also worth pointing out that the Rhetoric accords to literature the least honourable position in this hierarchy of pursuits, which is rhetorically effective: the addressee will have to appreciate the author’s undertaking for his sake even more. In the Ciris, on the contrary, the poet offers an apology for writing a mythological rather than philosophical poem, which, however, implies the same hierarchical relation between, generally speaking, literature and philosophy. As an interim summary, I would propose the following schematic outline of the Ciris’ conceptual background concerning the relation of poetry to other activities. In the most general terms, we find, on the one hand, the Greek opposition of poetry and philosophy and, on the other, the Roman one of otium and negotium. There seem to exist two ways in which the Greek scheme can be related to the Roman. First, in the traditional solution, both poetry and philosophy can be accommodated within the domain of otium. Second, in a more innovative solution offered by Catullus, philosophy can be matched against negotium as a serious occupation and poetry against otium as an unserious occupation. At first sight, what we see in the Ciris may appear to be a return from Catullus’ model to the traditional one. In fact, however, it implicitly effects an even more radical reassertion of the original Greek binary opposition (and in doing so it comes closer to Catullus), since political concerns are only of a marginal significance for the Ciris poet. Although it seems likely that the introductory etsi period of the Ciris is modelled on the beginning of the Rhetoric to Herennius, or on a similar opening of a

 Guenzi () . Contrast Lyne () .

1.4

31

prose work,³⁴ I would suggest that the primary object of polemic is rather to be sought in poetry. And we do find a view on the relation of political activity, philosophy and poetry, very similar to that expressed in the Rhetoric, in a fragment of Cicero’s De consulatu suo (fr. 10.71‒78 Courtney): haec adeo penitus cura uidere sagaci, otia qui studiis laeti tenuere decoris, inque Academia umbrifera nitidoque Lyceo fuderunt claras fecundi pectoris artis. e quibus ereptum primo iam a flore iuuentae te patria in media uirtutum mole locauit. tu tamen anxiferas curas requiete relaxans, quod patria uacat, id studiis nobisque sacrasti.

Cicero, addressed by the Muse Urania, is here presented as a politician in the first place, who nevertheless (note tamen) finds time (quod patria uacat) for philosophy (studiis) and poetry (nobis, i. e. the Muses). If Knox is right in connecting Cicero’s (short-lived) reputation as Rome’s leading poet, reported by Plutarch (Cic. 2.4), with the De consulatu suo rather than his juvenile poetry,³⁵ we have a striking example of a man who can easily excel in poetry despite being a full-time politician and orator and claiming also to have a vivid interest in philosophy (even if it is relegated to the status of a mere propaedeutics). Since no other Late Republican or Augustan poets could reasonably claim to equal Cicero as a man of the state, they had to renounce any pretension to excellence in politics. This is apparently one reason, though of course not the only one, why the Ciris poet claims to have abandoned his political career.³⁶ There are no close verbal parallels between Cicero’s fragment and the Ciris, but the image of Academia umbrifera may well have inspired that of the Epicurean Garden which florentis uiridi Sophiae complectitur umbra (4). Furthermore, the reference to Wisdom as quattuor antiquis heredibus edita consors (15), implying the four major philosophical schools of antiquity, may be an acknowledgement of Cicero’s preference for the Academy and the Lyceum.

 Peirano ()  calls attention to the fact that the (probable) addressees of Catullus  and the Ciris, Hortalus and Messalla, were both famous as orators: the point may be, I suggest, that they both would be able to recognise allusion to a treatise on rhetoric.  Knox () ‒; as he puts it (on pages ‒): “Since the reading public did not yet know Lucretius’ De rerum natura in , and had not seen much, if anything, of the Catullan libellus, Cicero’s De consulatu suo may well have appeared an impressive achievement.”  Which, of course, would not exactly be true if the Ciris is by Gallus.

32

1 The proem (lines 1‒100)

An indirect confirmation of the relevance of the Ciceronian passage for the Ciris is derived from the fact that it likewise seems to lie behind another poetic autobiography with which it features more tangible parallels, namely the socalled sphragis of Virgil’s Georgics (4.559‒566):³⁷ haec super aruorum cultu pecorumque canebam et super arboribus, Caesar dum magnus ad altum fulminat Euphraten bello uictorque uolentis per populos dat iura uiamque adfectat Olympo. illo Vergilium me tempore dulcis alebat Parthenope studiis florentem ignobilis oti, carmina qui lusi pastorum audaxque iuuenta, Tityre, te patulae cecini sub tegmine fagi.

Here we find an explicit reference not only to studia (shared by the Ciris), but also to otia. Like Cicero, Virgil combines philosophical and poetic pursuits, but, unlike him, refrains from politics, leaving it to Augustus. In contrast to Cicero, but like the Ciris poet, Virgil depicts himself as a student in an Epicurean school (Parthenope, implying Siro’s school in Naples). At the same time, it seems relevant to point out that the Virgilian passage is also indebted to Catullus 68.15‒26,³⁸ which, in turn, we have seen to be modelled on a programmatic epigram by Philodemus. Perhaps the most evident formal link between Catullus and Virgil is established by the collocation dulcis alebat, only attested at 68.24 (= 96) and Georgics 4.563. Virgil’s carmina qui lusi can thus be traced, through Catullus’ multa satis lusi, to Philodemus’ καὶ παίζειν ὅτε καιρός, ἐπαίξαμεν, and studiis florentem ignobilis oti, through hoc studium, to λωιτέρης φροντίδος; likewise audaxque iuuenta can be contrasted, through aetas florida uer ageret, with συνετῆς ἄγγελος ἡλικίης. In Virgil, however, there is already no apparent conflict between poetry and philosophy, but they combine together in opposition to politics.

1.5 Much of what follows will be concerned with how the Ciris proem constructs the relation between poetry and philosophy, or perhaps rather between mythological

 I discuss Virgil’s engagement with this fragment of De consulatu suo in greater detail elsewhere: see Kayachev (forthcoming-a). Cicero’s fragment has also been argued to have influenced other Virgilian contexts, both in the Georgics and the Aeneid, see Setaioli ().  Cf. Kayachev () .

1.5

33

poetry and philosophical poetry. We may start by pointing out that the opening period of the Ciris, which we have seen to be in an intricate dialogue with Catullus 65 and its subtexts, seems also to be indebted to the so-called second proem of Lucretius’ first book (1.922‒930): nec me animi fallit quam sint obscura; sed acri percussit thyrso laudis spes magna meum cor et simul incussit suauem mi in pectus amorem Musarum, quo nunc instinctus mente uigenti auia Pieridum peragro loca nullius ante trita solo. iuuat integros accedere fontis atque haurire iuuatque nouos decerpere flores insignemque meo capiti petere inde coronam, unde prius nulli uelarint tempora Musae.

Although the expressions they use do not always coincide, the Ciris and Lucretius certainly share a significant number of important motifs: ascent to the heights of (philosophical) inspiration unknown to others (8 placitum paucis ausa est ascendere collem, cf. auia Pieridum peragro loca nullius ante | trita solo), intellectual journey (mens … ascendere, cf. mente uigenti … peragro), striving for fame (laudis amore, cf. laudis spes), love for the Muses (blandum … amorem, cf. incussit suauem … amorem and, in a related context,³⁹ 1.19 incutiens blandum … amorem). We shall return to this parallel later again, whereas at the moment I propose to focus on a particular motif, that of the ascent to the heights of inspiration (given by either a philosophical figure such as Wisdom or the Muses), which is also, as we remember, the central point of analogy between the Ciris and the Tiresias section of Callimachus’ hymn. This motif is more fully developed in a later passage of the Ciris, which we find firmly grounded in the tradition of philosophical didactic (14‒17): si mihi iam summas Sapientia panderet arces, quattuor antiquis heredibus edita consors, unde hominum errores longe lateque per orbem despicere atque humilis possem contemnere curas…

In the first place, it unmistakably evokes a context from the opening of the second book of Lucretius’ poem (2.7‒14):⁴⁰

 Cf. Gale () .  As is, of course, well known, see e. g. Lyne ()  and Fowler () .

34

1 The proem (lines 1‒100)

sed nihil dulcius est, bene quam munita tenere edita doctrina sapientum templa serena, despicere unde queas alios passimque uidere errare atque uiam palantis quaerere uitae, certare ingenio, contendere nobilitate, noctes atque dies niti praestante labore ad summas emergere opes rerumque potiri. o miseras hominum mentes, o pectora caeca!

In turn, the Lucretian context has been linked with a fragment of Empedocles,⁴¹ which, as we shall see, may also be alluded to in the Ciris (fr. 3.6‒8): μηδέ σέ γ’ εὐδόξοιο βιήσεται ἄνθεα τιμῆς πρὸς θνητῶν ἀνελέσθαι, ἐφ’ ᾧ θ’ ὁσίης πλέον εἰπεῖν θάρσεϊ, καὶ τότε δὴ σοφίης ἐπ’ ἄκροισι †θοάζει†.

The ‘heights of wisdom’ (σοφίης ἐπ’ ἄκροισι), for which Empedocles is apparently encouraged to aspire,⁴² are certainly comparable to those occupied by the wise

 Bollack () . The fragment is notoriously full of difficulties, both textual and interpretative, and there is no universal agreement on how to solve them, cf. recently Trépanier () ‒ and Cerri (). Trépanier ()  even argues “that B  is really a collection of passages assembled on thematic grounds by Sextus’ source, perhaps in as many as four parts (B .‒; ‒; ; ‒).” If so, . (the line we are primarily interested in) becomes a single line torn out of context, which would leave too little evidence to relate it to either Lucretius or the Ciris. But Trépanier neglects one obvious argument for the unity of fr. .‒: though not line  itself, but both what precedes it and what follows it, are alluding to one and the same fragment of Parmenides (Emp. fr. . μηδέ σέ γ’ εὐδόξοιο βιήσεται cf. Parm. fr. . μηδέ σ’ ἔθος πολύπειρον ὁδὸν κατὰ τήνδε βιάσθω and Emp. fr. .‒ μήτε τιν’ ὄψιν ἔχων †πίστει† πλέον ἢ κατ’ ἀκουὴν | ἢ ἀκοὴν ἐρίδουπον ὑπὲρ τρανώματα γλώσσης cf. Parm. fr. .‒ νωμᾶν ἄσκοπον ὄμμα καὶ ἠχήεσσαν ἀκουὴν | καὶ γλῶσσαν). However, Trépanier seems right in punctuating after εἰπεῖν and taking θάρσει as an imperative, as well as in changing θοάζει to θόαζε. He also prefers Proclus’ τάδε τοι to Sextus’ τότε δὴ and interprets θόαζε transitively as ‘convey quickly τάδε upon the peaks of wisdom’ rather than in the dubious intransitive sense of ‘sit, throne upon’. This makes the whole line read: θάρσει καὶ τάδε τοι σοφίης ἐπ’ ἄκροισι θόαζε. I am not convinced that τάδε τοι is preferable to τότε δή, and in fact that either is true, so I would rather obelise it and take θόαζε intransitively as ‘make haste to’, the most natural sense in such a context, while interpreting ἐπί + dat. as a “pregnant construction” (LSJ, s.v., B.I..b): θάρσει καὶ †τότε δὴ† σοφίης ἐπ’ ἄκροισι θόαζε, ‘take heart, and so (?) make haste on to the peaks of wisdom and settle there’ (I italicise what LSJ adds by way of rendering the sense of ἐπί + dat.). I take the passage (partly because of the parallel in Parmenides) to be addressed to Empedocles by the Muse rather than by Empedocles to his disciple or the Muse, as it is sometimes understood: cf. A. Hardie () .  See the preceding note for my understanding of the fragment.

1.5

35

men in Lucretius, but also to the citadel of Wisdom onto which the narrator of the Ciris is ascending: we may note that the Ciris appears to reinterpret Empedocles’ σοφίη as a personification (Sapientia) rather than an abstract concept, as Lucretius did (doctrina). Empedocles’ fragment finds further points of contact in the opening period of the Ciris, where the narrator claims to have dared to climb the mountain of philosophy (8 ausa est ascendere collem, cf. – in Trépanier’s reconstruction – θάρσει καὶ … σοφίης ἐπ’ ἄκροισι θόαζε). Moreover, in contrast to Lucretius who wants to be honoured for his achievements in poetry rather than philosophy (1.922‒930), in both the Ciris (lines 1‒2) and Empedocles philosophical excellence is pointedly opposed to mundane glory, as honours (laudis and praemia, cf. εὐδόξοιο … ἄνθεα τιμῆς) bestowed by the mob (uulgi, cf. θνητῶν) can actually distract (iactatum, cf. βιήσεται) from wisdom (Sophiae, cf. σοφίης). The Ciris’ engagement with Empedocles (in addition to Lucretius) will find further confirmation in a later chapter where another possible case of ‘window’ allusion through Lucretius to Empedocles is treated (4§3). At the moment I would tentatively suggest that the Ciris is also alluding, likewise through Lucretius, to another Greek philosopher-poet, Parmenides. As argued recently,⁴³ Parmenides’ influence on Lucretius’ poetry may well be significantly more extensive than has generally been thought. One remarkably close Parmenidean parallel has been noted in the Lucretian passage quoted above (fr. 6.3‒7):⁴⁴ πρώτης γάρ σ’ ἀφ’ ὁδοῦ ταύτης διζήσιος 〈εἴργω〉, αὐτὰρ ἔπειτ’ ἀπὸ τῆς, ἣν δὴ βροτοὶ εἰδότες οὐδὲν πλάζονται δίκρανοι, ἀμηχανίη γὰρ ἐν αὐτῶν στήθεσιν ἰθύνει πλαγκτὸν νόον, οἱ δὲ φορεῦνται κωφοὶ ὁμῶς τυφλοί τε, τεθηπότες, ἄκριτα φῦλα.

Those wandering and erring in their search for a path in life whom Lucretius’ wise men have the pleasure of observing from their safe position are very much like Parmenides’ βροτοὶ εἰδότες οὐδέν who similarly wander along the false way of inquiry (2.10 errare atque uiam palantis quaerere uitae, cf. ὁδοῦ … διζήσιος, πλάζονται, φορεῦνται). Although, as is noted sometimes,⁴⁵ in form Lucretius’ condemnatory exclamation (2.14 o miseras hominum mentes, o pectora caeca!) more closely resembles a passage from Empedocles (fr. 124), in terms of content it surely is more indebted to Parmenides (cf. στήθεσιν, πλαγκτὸν  Rumpf ().  See Fowler ()  and , though he does not voice any opinion on the character of the relation.  Fowler () .

36

1 The proem (lines 1‒100)

νόον, κωφοὶ ὁμῶς τυφλοί τε).⁴⁶ As we have seen, precisely this Lucretian passage is imitated in the Ciris, and though the Ciris (hominum errores longe lateque per orbem) may not provide any specific signs that its author recognised Lucretius’ Parmenidean source, there is no reason either to think that he did not. In any event, and this might warn us against too much scepticism on the part of the Ciris, another Latin poet, Ovid, evidently did (Met. 15.150‒151): palantesque homines passim et rationis egentes | despectare procul trepidosque obitumque timentes, cf. βροτοὶ … πλάζονται (the latter at the same sedes as, and also phonetically similar to, palantes), εἰδότες οὐδὲν and ἄκριτα φῦλα (both at verse end like rationis egentes), τεθηπότες (comparable to trepidos in both sense and sound). Looking back at the parallel between the opening of the Ciris and the ‘second proem’ of Lucretius’ first book, we can now observe that two further shared motifs can be traced back to, or at least paralleled in, Parmenides. Let us start with the less certain case. The motif of the intellectual journey (or the flight of the mind) is more fully developed in another Lucretian passage, which praises Epicurus (1.72‒74): ergo uiuida uis animi peruicit et extra processit longe flammantia moenia mundi atque omne immensum peragrauit mente animoque.

The connection with the ‘second proem’ is unmistakable (note 1.925‒926 mente uigenti | auia Pieridum peragro loca) and significant, for it assimilates Lucretius’ poetic activity to Epicurus’ philosophical enquiry.⁴⁷ This parallel also indicates where the motif comes from: in its original form, it is a topos of philosophical rather than poetic discourse, and indeed in the Ciris we find it applied to the narrator qua philosopher rather than poet. As argued in earlier scholarship,⁴⁸ this motif is of Pythagorean origin, and in Lucretius’ laudes Epicuri it is used with such connotations in mind.⁴⁹ Parmenides’ philosophical revelation is likewise presented as taking place in the context of an intellectual journey, and again a Pythagorean background is the most probable explanation.⁵⁰ Perhaps it  Fowler () .  Cf. Gale () .  Edwards ().  Among other things, it has often been pointed out that the ‘praises of Epicurus’ contain an allusion to a fragment of Empedocles, generally taken to refer to Pythagoras, which likewise contains a variation on the motif of intellectual journey (note fr. . ὀρέξαιτο πραπίδεσσιν), see Edwards ()  n. .  See Burkert ().

1.5

37

would be unwise to force this rather general similarity between Parmenides and the Ciris too far, but it is not irrelevant for a fuller understanding of the literary context that the motif of the intellectual journey implies in the Ciris. The second point of resemblance linking the Ciris and Lucretius to Parmenides is more specific: the motif of the untrodden path (8 placitum paucis … collem, 1.926‒927 auia … loca nullius ante | trita solo). Again, Lucretius’ use of this motif has already been subjected to a detailed analysis by others, and likewise a Pythagorean provenance of the topos has been established.⁵¹ In more specific terms, two individual poetic parallels have been pointed out: one in the Aetia prologue (fr. 1.27‒28 κελεύθους | ἀτρίπτους) and the other in the proem of Parmenides’ poem (fr. 1.27 τήνδ’ ὁδόν, ἦ γὰρ ἀπ’ ἀνθρώπων ἐκτὸς πάτου ἐστίν), the former probably being modelled on the latter.⁵² In other words, a direct genetic line can arguably be drawn from Parmenides through Callimachus and then Lucretius to the Ciris. Finally, it is possible to point out a motif that seems to constitute a specific point of contact between the Ciris and Parmenides: the motif of the philosophical goddess receiving her disciple in her lofty abodes. The concise reference to Wisdom’s opening access to her citadel (si mihi iam summas Sapientia panderet arces) may indeed be an allusion to the detailed description of how Dike opens the gates of the dwelling place of Parmenides’ Goddess (fr. 1.11‒23).⁵³ It is to an extent irrelevant, and definitely impossible to ascertain beyond doubt, whether the Empedoclean and Parmenidean parallels we have noted are actually intended to be recognised as allusions. Whether they are or not, we can state that the picture of the didactic tradition produced in the Ciris is well-informed and convincing. Although Lucretius is certainly the primary target of reference, the Ciris poet is by no means an indiscriminating reader of his. It is indicative that Lucretius’ assimilation of inspiration to erotic desire, a motif that has no precedent in the didactic tradition (1.924 incussit suauem mi in pectus amorem is a pointed reminiscence of 1.19 incutiens blandum per pectora amorem, in its turn going back to h. Ven. 73 τοῖς ἐν στήθεσσι βάλ’ ἵμερον⁵⁴), is in the Ciris no longer associated with didactic poetry, but on the contrary with lighter genres (11 blandum liceat deponere amorem). By drawing an apparently clear-cut distinction between philosophical and mythological poetry, the Ciris seems also

 Knox ().  Cf. Kayachev ()  with further references.  Similarly, Verg. Georg. . accipiant caelique uias et sidera monstrent has been suggested to be an allusion to Parm. fr. . καί με θεὰ πρόφρων ὑπεδέξατο, see A. Hardie () . For other Parmenidean parallels in this Georgics context, see Erren () .  Flores () .

38

1 The proem (lines 1‒100)

to ignore Callimachus’ assimilation of his poetry to philosophy (which, however, is a false impression, as we shall soon see).

1.6 The first sentence of the Ciris, which we have mainly been dealing with so far, does not explain why the poet decides in favour of a mythological rather than a philosophical poem. In fact, it may give the impression that the reason is merely sentimental, that is either the poet’s own attachment to this sort of poetry, or the addressee’s predilection (though he has not been mentioned yet).⁵⁵ As we learn later, however, the reason is a more practical one: the poet still lacks expertise in philosophy (14‒45).⁵⁶ Yet we do get a glimpse of what the philosophical poem would be like, in an elaborate simile comparing it to the Panathenaic peplos (18‒41). This peplos image is of crucial importance for understanding the poetic programme of the Ciris, and we shall soon have a closer look at it; but first I propose to consider the structure of the first half of the proem (1‒53). In rough outlines, it can be divided into three parts: the first, announcing the general intention to compose a poem (1‒11); the second, explaining what sort of poem it will not be (12‒41); the third, finally specifying the poem’s subject (42‒ 53). As already mentioned, the Ciris is unlike many better-known ancient narrative poems in that it neither announces its subject right at the beginning (as, for example, the Iliad or the Aeneid do), nor plunges straight away into narration (as the Hecale or Catullus 64 do). It finds, however, a close structural parallel in the so-called Orphic Argonautica, whose proem displays a similar tripartite arrangement: first the narrator (Orpheus) invokes Apollo and prays for inspiration in general terms (1‒6); then he lists themes that he has already treated (7‒46); and finally he announces what his new composition will be about (47‒55). Moreover, just as in the Ciris there is a generic difference between the projected philosophical poem and the one we actually read, so there is a formal distinction between Orpheus’ previous compositions and the present one: previously he composed ‘mystical’ poetry in a state of divine possession, whereas now, freed from it, he can produce a conventional narrative poem. In fact, there is an even more specific point of resemblance. After describing what the hypothetical philosophical poem would be like, the Ciris poet proceeds  As Lyne ()  comments on  munus, “here it suggests too ‘gift’ (for Messalla): cf.  and .”  Though  promissa is ambiguous: was a poem of any sort promised, or specifically a mythological one?

1.6

39

to explain why he has to choose a different kind of poem and what his poem will be about (42‒50): sed quoniam ad tantas nunc primum nascimur artes, nunc primum teneros firmamus robore neruos, haec tamen interea quae possumus, in quibus aeui prima rudimenta et iuuenes exegimus annos, accipe dona meo multum uigilata labore promissa atque diu iam tandem 〈reddita uota〉 impia prodigiis ut quondam exterrita amoris Scylla nouos auium sublimis in aere coetus uiderit…

This passage manifests a formal similarity with the analogous passage of the Orphic Argonautica (47‒51): νῦν δ’, ἐπεὶ ἀερόφοιτος ἀπέπτατο δήιος οἶστρος ἡμέτερον δέμας ἐκπρολιπὼν εἰς οὐρανὸν εὐρύν, πεύσῃ ἀφ’ ἡμετέρης ἐνοπῆς ὁπόσα πρὶν ἔκευθον· ὥς ποτε Πιερίην Λειβήθρων τ’ ἄκρα κάρηνα ἡρώων τε καὶ ἡμιθέων πρόμος ἐξεπέρησε…

Both narrators start by explaining why the circumstances of the present composition (nunc, νῦν) are in some way extraordinary, then they invite their addressees to pay attention (accipe, πεύσῃ) to a poem that in ordinary circumstances they would not, or could not, produce (quae possumus, ὁπόσα πρὶν ἔκευθον), and finally they announce its subject by means of a subordinate clause (ut quondam, ὥς ποτε). The last detail seems especially remarkable, as normally the subject of an epic poem is named in a direct object rather than through a thatclause.⁵⁷ How is this similarity to be accounted for? It has been speculated that the Orphic Argonautica, usually considered a derivative late-antique composition, may be indebted not only to Greek poetry, but also to Latin. The conspicuous affinity of its proem with Latin examples of recusatio has been noted by Hunter;⁵⁸ though he did not posit a direct connection, one may be tempted to suspect that the Argonautica is here influenced by Latin poetry (perhaps specifically by the Ciris). While this possibility cannot be ruled out, an alternative explanation

 The use of a subordinate clause may be connected with the fact that both the Ciris and the Argonautica are addressed to a specific person (Messalla, Musaeus), which, again, is not typical for narrative poems.  Hunter () ‒.

40

1 The proem (lines 1‒100)

may be that both the Argonautica and the Ciris follow a Greek tradition that otherwise has not survived. In fact, the Ciris proem appears to echo another Greek epic proem, albeit for the most part lost, that of the Persica by Choerilus of Samos (SH 317):⁵⁹ ἆ μάκαρ, ὅστις ἔην κεῖνον χρόνον ἴδρις ἀοιδῆς, Μουσάων θεράπων, ὅτ’ ἀκήρατος ἦν ἔτι λειμών· νῦν δ’ ὅτε πάντα δέδασται, ἔχουσι δὲ πείρατα τέχναι, ὕστατοι ὥστε δρόμου καταλειπόμεθ’, οὐδέ πῃ ἔστι πάντῃ παπταίνοντα νεοζυγὲς ἅρμα πελάσσαι.

We do not know what the position and the context of this fragment in the proem was, and moreover the precise meaning of some terms within the fragment itself is not entirely clear (in particular in line 3). Yet it has some tangible points of contact with the Ciris. Perhaps the most evident one is the μακαρισμός the fragment begins with, which evokes an analogous passage in the Ciris (27‒28): felix illa dies, felix et dicitur annus, felices qui talem annum uidere diemque!

Formally speaking, both contexts exalt those who are related to a certain moment in time (κεῖνον χρόνον, talem annum): in Choerilus, it is the past; in the Ciris, every fourth year when the Great Panathenaea is celebrated. This may well appear to be a rather forced parallel, but there seems to be a point in it. Choerilus refers in that way to the famous poet(s) of the past long gone, presumably Homer in the first place, with whom he unfavourably contrasts himself. We may note that in the Ciris, too, the context belongs to the metaphorical description of a poem the author would write, but cannot. Moreover, Choerilus’ poem on the Persian war is reported to have been received so well that it was decreed to be performed together with the poems of Homer, most probably at the Panathenaic festival,⁶⁰ so that the μακαρισμός in the Ciris would actually apply to Choerilus in a very pregnant sense. The second point links Choerilus’ fragment not only with the Ciris, but also with the Orphic Argonautica, which may suggest that Choerilus’ lost proem had further structural analogies with the two other proems. This shared point is the poetologically significant opposition of the present moment to some other. Choerilus contrasts his modern-day poetry (νῦν δ’ ὅτε…) with that of the great poets of the past, Orpheus draws a distinction between the present composition

 On which see recently MacFarlane (), with further bibliography.  Cf. MacFarlane () ‒ n. .

1.6

41

(νῦν δ’ ἐπεί…) and his earlier poems, and in the Ciris the narrator apologises that what he can write now (nunc) is not as good as what he hopes to write in the future. Although the points of reference in Choerilus and the Ciris are the opposites of each other (the former refers to the past, the latter to the future), both poets can be said to be short of the ideal because they started their pursuit of the perfection of art too late (ἔχουσι δὲ πείρατα τέχναι, | ὕστατοι ὥστε δρόμου καταλειπόμεθ〈α〉, ad tantas nunc primum nascimur artes). In other words, Choerilus is losing the contest to those who started earlier, the Ciris poet to his own older self. This brings us to the third point, which Choerilus likewise shares with both the Ciris and the Orphic Argonautica: Choerilus, too, announces the subject of his poem in a subordinate clause rather than through a direct object (SH 316): ἥγεό μοι λόγον ἄλλον, ὅπως ᾿Aσίης ἀπὸ γαίης ἦλθεν ἐς Εὐρώπην πόλεμος μέγας.

This fragment is often assumed to be the very opening of Choerilus’ epic because it is cited by Aristotle (Rh. 1415a) together with the first lines of the Iliad and the Odyssey, but that conclusion is hardly compelling. It seems no less conceivable that SH 316 stood after the programmatic proem – to which, of course, SH 317 would have belonged – at the beginning of the narrative proper.⁶¹ In this case, the position of SH 316 would be roughly equivalent to that of the passage announcing the poem’s topic in both the Orphic Argonautica and the Ciris. There is one more potential point of contact with the Ciris. Choerilus is envious of those who worshipped the Muses at the time when the meadow was yet intact (Μουσάων θεράπων, ὅτ’ ἀκήρατος ἦν ἔτι λειμών). We do not know how, and whether at all, these two metaphors were connected with the fragment’s wider context. But we find in the Ciris a combination of roughly the same two metaphors at the closure of the proem (92‒100), where the narrator presents himself as a priest of the Muses and places their temple in the centre of a blooming meadow. Later on we shall return to this passage and consider in what way it may be refracting the poetological implications of Choerilus’ fragment. As has been demonstrated by Hollis, to the Augustan poets Choerilus is likely to have been not as obscure an author as he is to us.⁶² I would further suggest that there may be a particular reason for his presence in the Ciris. We are told in an epigram that Χοιρίλον Εὐφορίων εἶχε διὰ στόματος (AP 11.218.2):⁶³ if the Ciris

 This order is accepted in Colace’s () edition; cf. also MacFarlane () ‒ n. .  Hollis ().  Huxley () ‒ offers some interesting comments.

42

1 The proem (lines 1‒100)

is indeed by Gallus who in his turn can be identified as a cantor Euphorionis, it is only natural that echoes of Choerilus should be found in the Ciris.

1.7 We have seen how the Ciris patterns the fundamental opposition of philosophical and mythological poetry against a number of subtexts. At the same time, some subtexts, first of all Philodemus and Callimachus, would rather call into question whether this opposition is as impenetrable as the narrator may want us to believe. Let us now have a closer look at the Panathenaic peplos with which the hypothetical philosophical poem is compared (21‒26, 35): sed magno intexens, si fas est dicere, peplo, qualis Erectheis olim portatur Athenis, debita cum castae soluuntur uota Mineruae tardaque confecto redeunt quinquennia lustro, cum leuis alterno Zephyrus concrebuit Euro et prono grauidum prouexit pondere currum. … tale deae uelum sollemni tempore portant.

Lyne had doubts whether the Panathenaic peplos is an appropriate image for a philosophical poem: “A peplos embroidered with the martial deeds of Athena (29‒34) would most naturally symbolise a laudatory epic”; which led him to suspecting that “this attractive ecphrasis was already available; it has been borrowed.”⁶⁴ Lyne’s misgivings have partly been answered by Faber, who connected the peplos of the Ciris with the tradition of cosmological ecphrastic images in other Greek and Latin texts.⁶⁵ One particularly relevant parallel suggested by Faber, though only in general terms, is Jason’s cloak in Apollonius’ Argonautica, a gift from Athena (1.721‒724, 768):⁶⁶ αὐτὰρ ὅγ’ ἀμφ’ ὤμοισι, θεᾶς Ἰτωνίδος ἔργον, δίπλακα πορφυρέην περονήσατο, τήν οἱ ὄπασσε Παλλάς, ὅτε πρῶτον δρυόχους ἐπεβάλλετο νηὸς ᾿Aργοῦς καὶ κανόνεσσι δάε ζυγὰ μετρήσασθαι. … τοῖ’ ἄρα δῶρα θεᾶς Ἰτωνίδος ἦεν ᾿Aθήνης.

 Lyne () .  Faber ().  Faber () ; cf. Nagy () ‒.

1.7

43

It seems possible to argue that Apollonius’ ecphrasis is actually a model of, not just a parallel to, the peplos in the Ciris. To start with their appearance, the ‘purple suns’ embroidered on the peplos (37 purpureos inter soles) find a certain correspondence in the ‘rising sun’ with which the colour of Jason’s cloak is compared (1.725‒726 τῆς μὲν ῥηίτερόν κεν ἐς ἠέλιον ἀνιόντα | ὄσσε βάλοις ἢ κεῖνο μεταβλέψειας ἔρευθος).⁶⁷ Both Jason’s cloak and the Panathenaic peplos are gifts, either from or for, Athena (cf. τοῖ’ ἄρα δῶρα θεᾶς and tale deae uelum). Both the cloak and the peplos are presented, either by or to Athena, on an occasion in some way involving a ship (ὅτε πρῶτον δρυόχους ἐπεβάλλετο νηός, cum leuis alterno Zephyrus concrebuit Euro | et prono grauidum prouexit pondere currum). In fact, the wheeled ship (currum) on which the peplos was carried during the Panathenaic procession is arguably meant to evoke Catullus’ mention of the construction of the Argo (64.9‒10 ipsa leui fecit uolitantem flamine currum, | pinea coniungens inflexae texta carinae), a context clearly linked with the Apollonian passage.⁶⁸ It is certainly not irrelevant that Apollonius’ ecphrasis was interpreted as an allegorical cosmic image by an ancient scholiast (on 1.763):⁶⁹ ζητητέον δὲ τί βούλεται αὐτῷ ταῦτα τὰ ποικίλματα. καὶ ἐροῦμεν, ὅτι ὁ ποιητὴς διὰ τῆς χλαμύδος οὐδὲν ἕτερον ἢ τὴν κοσμικὴν τάξιν καὶ τὰς τῶν ἀνθρώπων πράξεις φησί. καὶ πρῶτον μὲν διὰ τοῦ κεραυνοῦ καὶ τῶν Κυκλώπων θεόν τινα καὶ θείαν φύσιν ἀλληγορεῖ. […] δῶρον δέ φησιν εἶναι τὴν χλαμύδα τῆς ᾿Aθηνᾶς, ἐπειδὴ ὅ τε κόσμος ὑπὸ τῆς θείας φρονήσεως γέγονε, τῶν τε ἐν αὐτῷ ὑπὸ τῶν ἀνθρώπων πραττομένων οὐδὲν ἄνευ φρονήσεως ὀρθῶς γένοιτ’ ἄν.

This interpretation is no doubt confirmed, in general if not in all details, by the fact that the first scene on the cloak (1.730‒731 ἐν μὲν ἔσαν Κύκλωπες ἐπ’ ἀφθίτῳ ἥμενοι ἔργῳ, | Ζηνὶ κεραυνὸν ἄνακτι πονεύμενοι) picks up the narrative of the cosmogonic song of Orpheus (1.509‒510 οἱ δέ μιν οὔπω | γηγενέες Κύκλωπες ἐκαρτύναντο κεραυνῷ).⁷⁰ It seems interesting to point out that the scholiast explains Athena as an allegory of intelligence, φρόνησις, which can be compared

 On the level of content, the first panel of Jason’s cloak that pictures the Cyclops forging Zeus’ thunderbolt (.‒) can be compared with  additur aurata deiectus cuspide Typhon, with cuspide arguably referring to Zeus’ thunderbolt rather than Athena’s spear, cf. Eur. Hec. ‒  Τιτάνων γενεὰν | τὰν Ζεὺς ἀμφιπύρῳ κοιμί|ζει φλογμῷ Κρονίδας (likewise decoration of the Panathenaic peplos).  Note that both Apollonius (κανόνεσσι) and Catullus (texta) use terms that can be applied to both shipbuilding and weaving.  Cf. Levin () ‒, Shapiro () , Bulloch ()  n. , Faber () .  Cf. George () , Nelis () , Hunter (a) ‒.

44

1 The proem (lines 1‒100)

with the figure of Sapientia in the Ciris (it can be added perhaps that 14 summas Sapientia panderet arces may be alluding to Catullus 64.8 retinens in summis urbibus arces, referring to Athena). Much as the embroidery of the Panathenaic peplos is inspired by the martial deeds of Athena, so a philosophical poem needs to be inspired by Wisdom. Perhaps most importantly, it can be argued that Jason’s cloak is itself meant to evoke the Panathenaic peplos,⁷¹ so that the Ciris only makes explicit what was already implied in Apollonius. Among the numerous Homeric models underlying Jason’s cloak, I would call special attention to one that describes the offering of a peplos to Athena by the Trojan women, which is an obvious analogue of what happened at the Panathenaea (Il. 6.288‒296):⁷² αὐτὴ δ’ ἐς θάλαμον κατεβήσετο κηώεντα, ἔνθ’ ἔσαν οἱ πέπλοι παμποίκιλοι, ἔργα γυναικῶν Σιδονίων, τὰς αὐτὸς ᾿Aλέξανδρος θεοειδὴς ἤγαγε Σιδονίηθεν, ἐπιπλὼς εὐρέα πόντον, τὴν ὁδὸν ἣν Ἑλένην περ ἀνήγαγεν εὐπατέρειαν. τῶν ἕν’ ἀειραμένη Ἑκάβη φέρε δῶρον ᾿Aθήνῃ, ὃς κάλλιστος ἔην ποικίλμασιν ἠδὲ μέγιστος, ἀστὴρ δ’ ὣς ἀπέλαμπεν· ἔκειτο δὲ νείατος ἄλλων. βῆ δ’ ἰέναι, πολλαὶ δὲ μετεσσεύοντο γεραιαί.

Apollonius’ points of contact with the Homeric passage may not be particularly striking, but they are noticeable. Both Jason’s cloak and the peplos are first described as handiwork (θεᾶς Ἰτωνίδος ἔργον, cf. ἔργα γυναικῶν) and then as a gift, either from or for Athena (δῶρα … ᾿Aθήνης, cf. δῶρον ᾿Aθήνῃ). Both contexts display a star simile (1.774 βῆ δ’ ἴμεναι προτὶ ἄστυ, φαεινῷ ἀστέρι ἶσος, of Jason wearing the cloak, cf. ἀστὴρ δ’ ὣς ἀπέλαμπεν and note βῆ δ’ ἰέναι beginning the next line). If, therefore, Jason’s cloak is indeed a subtext of the Panathenaic peplos in the Ciris, the use of the latter as an image for a philosophical poem becomes quite understandable, but at the same time somewhat problematic. It is true, as we have seen, that the Apollonian ecphrasis can be read, and actually was in antiquity, as an allegorical image of the universe; accordingly, it is no surprise that the Panathenaic peplos should have a similar cosmological significance in the Ciris. Yet there is an intrinsic paradox in that it should stand for a separate

 Cf. Nagy () .  As Nagy () ‒ is right to emphasise, the Athenian audience of the Iliad (officially performed at the Panathenaea) would no doubt see in this Homeric scene an allusion to the Panathenaic procession.

1.7

45

philosophical poem that is generically opposed to mythological epic, for in Apollonius, of course, the ecphrasis is inseparably incorporated in the story. We may, in fact, remember that epic, and especially Homeric epic, was considered in antiquity the most universal genre, whose subject is not only the deeds of men, but the entire universe.⁷³ A cosmological ecphrasis like that of Jason’s cloak only brings this universal dimension out, only puts a particular human story within the framework of the whole of world history. Such an ecphrasis can thus be said to contain within itself, in a sense, the whole poem it is part of; it becomes an epic in miniature, a matrix from which the poem can be said to develop. We find this notion realised in another Homeric model of the Apollonian ecphrasis, a passage that pictures Helen weaving a cloth decorated with the figures of the fighting Trojans and Greeks (3.125‒131):⁷⁴ τὴν δ’ εὗρ’ἐν μεγάρῳ, ἣ δὲ μέγαν ἱστὸν ὕφαινε, δίπλακα μαρμαρέην, πολέας δ’ ἐνέπασσεν ἀέθλους Τρώων θ’ ἱπποδάμων καὶ ᾿Aχαιῶν χαλκοχιτώνων, οὕς ἕθεν εἵνεκ’ ἔπασχον ὑπ’ Ἄρηος παλαμάων. ἀγχοῦ δ’ ἱσταμένη προσέφη πόδας ὠκέα Ἶρις· δεῦρ’ ἴθι, νύμφα φίλη, ἵνα θέσκελα ἔργα ἴδηαι Τρώων θ’ ἱπποδάμων καὶ ᾿Aχαιῶν χαλκοχιτώνων…

As a scholiast comments on this cloth (bT on 3.126‒127), ἀξιόχρεων ἀρχέτυπον ἀνέπλασεν ὁ ποιητὴς τῆς ἰδίας ποιήσεως. The equivalence of the events depicted on the cloth and those taking place in reality (that is, in the poem) is visualised by an exact repetition of a whole line: Τρώων θ’ ἱπποδάμων καὶ ᾿Aχαιῶν χαλκοχιτώνων.

 Cf. e. g. P. Hardie () ‒.  Nagy ()  argues that of the two epithets of δίπλακα, μαρμαρέην and πορφυρέην, attested as variants at both Il. . and ., the former “refers to the luminosity of the purple” and the latter “simply to the color,” and suggests that μαρμαρέην presupposes an Athenian audience which, recognising the parallel with the Panathenaic peplos, would infer that the colour is purple, whereas πορφυρέην aims at a wider Hellenistic audience which needs the colour to be specified. Although van Thiel ()  seems right in supposing that μαρμαρέην is the original reading at . and πορφυρέην at . and that the variants have originated as cross-references, it is likely that Alexandrian scholars did see the two epithets as implying the same colour. Whilst apparently preferring πορφυρέην from ., Apollonius may be hinting at μαρμαρέην in . with .‒ τῆς μὲν ῥηίτερόν κεν ἐς ἠέλιον ἀνιόντα | ὄσσε βάλοις ἢ κεῖνο μεταβλέψειας ἔρευθος. Similarly, I would tentatively suggest, Hor. Ars ‒ purpureus, late qui splendeat, unus et alter | adsuitur pannus may be meant to allude to both Homeric epithets.

46

1 The proem (lines 1‒100)

It is difficult to find positive proofs that either of the two Homeric contexts is relevant for the Panathenaic peplos of the Ciris. If, however, we accept a pre-Virgilian dating of the Ciris, we may find a suggestive argument in the description of Dido’s murals in the Aeneid (1.456‒493), which can be argued to contain a ‘window’ allusion through the Ciris to Homer. Thomas has convincingly shown that the central position of the scene of peplophoria on the murals (1.479‒481), a scene that seems out of place among those of combat, is intended to stress the ecphrastic potential inherent – through its association with the Panathenaic peplos, among other things – in the Homeric model.⁷⁵ As he has also observed, “in Iliad 3 Helen is at work weaving precisely the objects which Virgil was to place on Dido’s mural.”⁷⁶ To be more specific, the opening of the Virgilian ecphrasis – uidet Iliacas ex ordine pugnas (1.456) – can be matched against the summarising description of Helen’s patterns: πολέας δ’ ἐνέπασσεν ἀέθλους. But it also happens that the account of the scenes depicted on the Panathenaic peplos in the Ciris begins with a line – ergo Palladiae texuntur in ordine pugnae (29) – that features the same collocation at its end as the Virgilian line, a collocation unattested elsewhere in Latin poetry. Unless we discount this parallel as coincidental, I would tentatively suggest that it is the Ciris line that can be linked with πολέας δ’ ἐνέπασσεν ἀέθλους (Τρώων θ’ ἱπποδάμων καὶ ᾿Aχαιῶν χαλκοχιτώνων) in the first place, whereas Virgil is rather alluding – in order to reproduce the Homeric repetitio cum variatione – to ἵνα θέσκελα ἔργα ἴδηαι (Τρώων θ’ ἱπποδάμων καὶ ᾿Aχαιῶν χαλκοχιτώνων): texuntur would thus be rendering ἐνέπασσεν and Virgil’s uidet translating ἴδηαι. In fact, we can see this remarkable Homeric repetition imitated, in a way, within the Ciris itself. After describing the scenes depicted on the Panathenaic peplos, the narrator proceeds (36‒39): tali te uellem, iuuenum doctissime, ritu purpureos inter soles et candida lunae sidera, caeruleis orbem pulsantia bigis, naturae rerum magnis intexere chartis.

We may first observe that the two middle lines are clearly intended – note intexere – as an iconic representation of the interlacing of threads of different colours: purple, white, blue.⁷⁷ It has been suggested that the pervasive metaphor of weaving applied to poetic composition (9 detexere, 21 intexens, 29 texuntur [literally],

 Thomas () ‒.  Thomas () .  On similar instances of ‘mimetic’ word-order in Ovid, cf. Lateiner ().

1.7

47

39 intexere, 100 praetexite) may be related in the Ciris to Lucretius’ practice of speaking – “[i]n the tradition of Greek materialist philosophy, which since Leucippus and Democritus considered physical reality a sumplokē” – of “the ‘atomic structure’ as a textura, a textum, or a textus.”⁷⁸ The metaphor of weaving therefore would be particularly suitable for a poem on the universe; in the Ciris, however, it is used not only of the projected philosophical poem, but also of the mythological poem we actually read. This appears to suggest that, much as different objects in the universe consist of a limited number of varieties of atoms, so poems of different genres are composed of the same words and, ultimately, letters (ἐκ τῶν αὐτῶν γὰρ τραγῳδία καὶ κωμῳδία γίνεται γραμμάτων, as Aristotle, GC 315b, puts it).⁷⁹ Our intuition is confirmed as we find at the beginning of the narrative part a distinct internal echo of the ‘iconic’ passage quoted above (101‒ 103):⁸⁰ sunt Pandioniis uicinae sedibus urbes Actaeos inter colles et candida Thesei purpureis late ridentia litora conchis.

The metaphor of weaving is activated by praetexite in the last line of the proem (100), and here two of the three colour terms are repeated (in the same metrical position). We can see how soles turns into colles, by the change of only a single consonant, more or less; taking the hint, I would further suggest that urbes is likewise meant to evoke orbem, thus establishing a fundamental analogy between the cosmic and the human levels (that is, between philosophical and mythological poetry).⁸¹ If my argument that the image of the Panathenaic peplos belongs to the tradition of cosmological ecphrases is valid, we are thus invited to reconsider the relation between philosophical and mythological poetry – or, more generally, between philosophy and poetry – that is constructed in the Ciris. Instead of a clear-

 Scheid and Svenbro ()  and cf. ‒.  On such an ‘atomist’ poetics, cf. Armstrong (). It seems relevant to emphasise that the analogy can be applied in both directions: whereas the atomists would use it to explain the atomic structure of the material world (as Lucretius does, see e. g. Maltby []), the Ciris poet applies it to assert the fundamental unity underlying different poetic genres (which is a Philodemean idea, cf. Armstrong [] ).  The reference is further confirmed by the parallel between ‒ praecipue nostro nunc aspirate labori | atque nouum aeterno praetexite honore uolumen and ‒ aeterno ut Sophiae coniunctum carmine nomen | nostra tuum senibus loqueretur pagina saeclis.  In a similar manner, Ciris  longe lateque per orbem appears to be directly related to Verg. Aen. . longe lateque per urbes, the collocation longe lateque per being unattested elsewhere.

48

1 The proem (lines 1‒100)

cut opposition of two kinds of poetry that results from a literal reading of the proem, the Ciris poet, I would contend, implicitly asserts rather than denies a certain compatibility between (mythological) poetry and philosophy. The way the description of the peplos resonates in the opening of the narrative part suggests that even mythological poetry can be informed by philosophy: a geographical ecphrasis (inter colles et candida) is literally patterned on an astronomical one (inter soles et candida). Indeed, as we shall regularly observe in later chapters, Lucretius’ philosophical poem is constantly present in the background of the Ciris, not only in the proem. Yet this relationship – of interaction rather than opposition – is not a straightforward one either. Although at first sight it may appear that the celestial takes precedence over the terrestrial, this notion is subtly subverted: the ‘purple suns’ (purpureos … soles), which are only ornaments on the peplos, turn out to be in a sense derivative of the ‘purple molluscs’ (purpureis … conchis), the source of the purple dye.⁸²

1.8 If the first half of the proem can be said to define the forthcoming composition in terms of genre, the second half specifies its subject. The Ciris poet warns his addressee that the Megarian princess Scylla, who will be the heroine of the poem, should not be thought to have transformed into the Homeric monster, as is sometimes claimed (54‒61); and then lists stories about other girls who have better claims to be identified with her monstrous namesake (62‒91). Little could be gained from a discussion of these stories, since in most cases we do not know the sources the Ciris refers to. But the passage that introduces this digression on alternative versions is worth a few comments (54‒57): complures illam magni, Messalla, poetae (nam uerum fateamur: amat Polyhymnia uerum) longe alia perhibent mutatam membra figura Scyllaeum monstro saxum infestasse uoraci.

 Lyne ()  believes “that purpureis denotes ‘shining’ ”; yet Edgeworth () seems right to argue that the only lexical meaning purpureus can have is ‘purple’. The closest extant parallel for purpureis … conchis is suggestive (Nonn. D. .): πορφυρέῳ πεπαλαγμένα φάρεα κόχλῳ.

1.8

49

A fragment of Euphorion’s Hyacinthus has often been adduced as a comparandum (fr. 44 Lightfoot):⁸³ πορφυρέη ὑάκινθε, σὲ μὲν μία φῆμις ἀοιδῶν Ῥοιτείῃς ἀμάθοισι δεδουπότος Αἰακίδαο εἴαρος ἀντέλλειν τεὰ γράμματα κωκύουσαν.

It is clear that Euphorion was going to mention alternative stories, too (note the allusion to Od. 22.376 σύ τε καὶ πολύφημος ἀοιδός). The link with Euphorion finds support in the well-known parallel passage in Virgil’s sixth eclogue (6.74‒76 quam fama secuta est | candida succinctam latrantibus inguina monstris | Dulichias uexasse rates: lines 6.75‒77 almost entirely coincide with Ciris 59‒61), where fama with a dependent accusative and infinitive construction (quam … uexasse) is an exact equivalent of φῆμις likewise governing an accusative and infinitive construction (σὲ … ἀντέλλειν).⁸⁴ Whether or not Virgil is using the Ciris rather than the other way round, either author would have recognised his predecessor’s subtext in Euphorion. Another possible subtext can be found in Callimachus’ hymn to Delos, where similarly the subject is specified first in general terms and then a particular story is chosen among many others (4.28‒29): εἰ δὲ λίην πολέες σε περιτροχόωσιν ἀοιδαί, ποίῃ ἐνιπλέξω σε; τί τοι θυμῆρες ἀκοῦσαι;

To begin with, the metaphor of weaving finds an interesting parallel earlier in the proem: while Callimachus still has to choose into which song he will weave his subject (ποίῃ ἐνιπλέξω σε), the Ciris poet already knows into what kind of poem he would like to weave his addressee (36‒39 te uellem … naturae rerum magnis intexere chartis, cf. 18‒21). More to the point, the Callimachean subtext draws attention to the etymological wordplay on complures (rendering πολέες) and Polyhymnia ⁸⁵ (note further that uerum fateri is virtually a translation of ἐτυμολογεῖν⁸⁶). Thus, despite Lyne’s bewilderment (“There seems no particular reason for her, or any Muse, being represented as thus enamoured of truth – but no particular reason why not”⁸⁷), there is a particular reason to choose Poly-

 Skutsch () , Lyne () , Hollis () .  See Skutsch () .  For (possible) instances of etymological wordplay on Polyhymnia’s name in Greek poetry, see A. Hardie () ‒.  On etymological signposts, cf. Maltby () , Cairns () ‒.  Lyne () .

50

1 The proem (lines 1‒100)

hymnia, that is for the sake of wordplay. But there is also a “particular reason why not,” or at least why this choice is rather problematic. Since Polyhymnia is thus a personification of πολέες ἀοιδαί (~ ὕμνοι), or a patron of complures poetae, whose assertions are censured as unreliable (cf. also 89‒90 quidquid et ut quisque est tali de clade locutus, | somnia sunt), the statement amat Polyhymnia uerum sounds controversial, to say the least;⁸⁸ and even more so in the light of the famous passage from Hesiod’s Theogony (27‒28) where the Muses claim authority over both truth and falsehood. Also relevant may be a further Callimachean parallel, a passage in the hymn to Zeus that similarly criticises poetic tradition (1.60): δηναιοὶ δ’ οὐ πάμπαν ἀληθέες ἦσαν ἀοιδοί. It may be difficult to decide whether uerum and poetae in the Ciris were actually prompted by ἀληθέες and ἀοιδοί (rather than ἀοιδαί in the other hymn), but it seems likely, as argued by Brown, that this line lies behind Lucretius’ analogous attack on poetic tradition (2.600‒601): hanc ueteres Graium docti cecinere poetae | sedibus in curru biiugos agitare leones. ⁸⁹ The rare occurrences of poeta in later Latin hexameters (mostly didactic) betray Lucretius’ influence,⁹⁰ and it seems legitimate to take the Ciris passage as a ‘window’ allusion through Lucretius to Callimachus. A final argument in support both of the Callimachean allusion in the Ciris and of the paradoxical implication of the choice of Polyhymnia can be derived from Ovid. The fifth book of the Fasti opens with a passage that combines an allusion to Callimachus with a mention of Polyhymnia (5.1‒10): quaeritis unde putem Maio data nomina mensi? non satis est liquido cognita causa mihi. ut stat et incertus qua sit sibi nescit eundum, cum uidet ex omni parte, uiator, iter, sic, quia posse datur diuersas reddere causas, qua ferar ignoro, copiaque ipsa nocet. dicite, quae fontes Aganippidos Hippocrenes, grata Medusaei signa, tenetis, equi. dissensere deae; quarum Polyhymnia coepit prima (silent aliae, dictaque mente notant).

 Note that Plato (Symp. d–e) associates Polyhymnia with πάνδημος ἔρως, in opposition to Urania who is linked with οὐράνιος ἔρως; cf. A. Hardie () .  Brown () .  Ovid (in Met.), Lucan, Valerius Flaccus, Silius Italicus and Statius do not use it at all. Apart from five instances in the eclogues, Virgil only has one use in the Georgics, clearly alluding to the Lucretian passage (.‒): quorum Grai meminere poetae, | Martis equi biiuges et magni currus Achilli. Likewise, the only context in Germanicus is indebted to Lucretius (): haec ego non primus, ueteres cecinere poetae. Manilius’ only use (.) may be irrelevant.

1.9

51

Ovid is confronted by the same problem of choice among multiple versions (lines 5.5‒6, cf. εἰ δὲ λίην πολέες…), but unlike Callimachus, instead of choosing a single story, he introduces three Muses to present three alternative etymologies of the name of the month May. The first Muse to give her account is Polyhymnia, the patron of ‘many songs’, and this coincidence is especially remarkable since she is only mentioned on very few occasions elsewhere in Latin poetry.⁹¹ In a later chapter (5§4) we shall note and discuss further parallels linking the Ciris, Callimachus’ hymn to Delos, and the first episode of the fifth book of Ovid’s Fasti. At the moment it will be enough to point out, in anticipation of the next section, the Ciris poet’s awareness of the complexity and diversity of poetic tradition, which he manifests not only by explicitly acknowledging it, but also by means of allusion.

1.9 The proem concludes with an elaborate invocation of the Muses, which combines two poetological motifs we saw in Choerilus: the servant of the Muses and the virgin meadow (92‒100): quare, quae cantus meditanti mittere doctos magna mihi cupido tribuistis praemia, diuae Pierides, quarum castos altaria postis munere saepe meo inficiunt foribusque hyacinthi(1) deponunt flores aut suaue rubens narcissus(2) aut crocus(3) alterna coniungens lilia(4) caltha(x) sparsaque liminibus floret rosa(5), nunc age, diuae, praecipue nostro nunc aspirate labori atque nouum aeterno praetexite honore uolumen.

As we remember, Choerilus praises the past when the meadow of poetry was yet intact, in contrast to his contemporary situation ὅτε πάντα δέδασται (whatever that may mean). We do not know what was his answer to the challenge of writing poetry in this position of belatedness; but we can see what solution is offered by the Ciris poet when he is confronted by the same problem. In contrast to Choerilus, he presents himself – on the surface level – as a priest of the Muses who worships at the altar placed in the centre of an idyllic meadow; yet on the

 Apart from one occurrence in the Ciris and two in Ovid (the second one comes later in this same episode, Fast. .), there is only one context in Horace (Carm. ..) and one in Martial (..).

52

1 The proem (lines 1‒100)

level of subtexts this meadow is not as virgin as it may look, for it is pointedly modelled on a long poetic tradition of flower catalogues. The most immediate model comes from Moschus’ Europa (65‒71): τῶν ἣ μὲν νάρκισσον(2) ἐύπνοον, ἣ δ’ ὑάκινθον(1), ἣ δ’ ἴον(4), ἣ δ’ ἕρπυλλον(x) ἀπαίνυτο· πολλὰ δ’ ἔραζε λειμώνων ἐαροτρεφέων θαλέθεσκε πέτηλα. αἳ δ’ αὖτε ξανθοῖο κρόκου(3) θυόεσσαν ἔθειραν δρέπτον ἐριδμαίνουσαι· ἀτὰρ μέσσῃσιν ἄνασσα ἀγλαΐην πυρσοῖο ῥόδου(5) χείρεσσι λέγουσα οἷά περ ἐν Χαρίτεσσι διέπρεπεν ᾿Aφρογένεια.

In its turn, Moschus’ catalogue is modelled on that in the Homeric hymn to Demeter (6‒9 and 425‒428): ἄνθεά τ’ αἰνυμένην ῥόδα(5) καὶ κρόκον(3) ἠδ’ ἴα(4) καλὰ λειμῶν’ ἂμ μαλακὸν καὶ ἀγαλλίδας(x) ἠδ’ ὑάκινθον(1) νάρκισσόν(2) θ’, ὃν φῦσε δόλον καλυκώπιδι κούρῃ Γαῖα… … παίζομεν ἠδ’ ἄνθεα δρέπομεν χείρεσσ’ ἐρόεντα, μίγδα κρόκον(3) τ’ ἀγανὸν καὶ ἀγαλλίδας(x) ἠδ’ ὑάκινθον(1) καὶ ῥοδέας(5) κάλυκας καὶ λείρια(4), θαῦμα ἰδέσθαι, νάρκισσόν(2) θ’ ὃν ἔφυσ’ ὥς περ κρόκον εὐρεῖα χθών.

And the two Greek contexts have a third potentially relevant parallel in a fragment of the Cypria (fr. 4 Bernabé): εἵματα μὲν χροῒ ἕστο, τά οἱ Χάριτές τε καὶ Ὧραι ποίησαν καὶ ἔβαψαν ἐν ἄνθεσιν εἰαρινοῖσιν, οἷα φοροῦσ’ ὧραι, ἔν τε κρόκῳ(3), ἔν θ’ ὑακίνθῳ(1), ἔν τε ἴῳ(4) θαλέθοντι ῥόδου(5) τ’ ἐνὶ ἄνθεϊ καλῷ ἡδέι νεκταρέῳ, ἔν τ’ ἀμβροσίαις καλύκεσσιν αἰθέσι ναρκίσσου(2) καλλιπνόου. ὧδ’ ᾿Aφροδίτη ὥραις παντοίαις τεθυωμένα εἵματα ἕστο.

We may start by quoting Campbell’s comments on the Moschus passage, which will be relevant for the Ciris as well: In Homeric Hymn to Demeter Persephone’s abduction is preceded by a list of flowers (6‒8) specifying five out of the six mentioned here: instead of the rare and difficult ἀγαλλίδες of the hymn […] we get ἕρπυλλος […]. The five figure also in a flower-catalogue in Cypria EGF

1.9

53

F4 […]. It cannot be a coincidence that Moschus has exactly reversed the order of the five flower-types offered by the hymn.⁹²

In fact, there may be a more specific reason for replacing ἀγαλλίδες with ἕρπυλλος, which Campbell does not note explicitly, namely that ἀγαλλίδες are the only sort of flowers not mentioned in the Cypria fragment and can accordingly be deemed less traditional.⁹³ That Moschus does take notice of the Cypria fragment seems to be confirmed by the comparison of Europa with Aphrodite among the Graces. Now, turning back to the Ciris, we can note that in addition to the five ‘traditional’ flower species common to the Cypria, the hymn, and Moschus (all of which are listed in the Ciris), and instead of the facultative ἀγαλλίδες or ἕρπυλλος, the Latin poem has its own innovation, caltha. ⁹⁴ Obviously, this is not likely to be a coincidence, and it is a natural inference that the Ciris poet should have known at least two of the three earlier texts. Thus, in contrast to Choerilus’ ideal of a wild meadow, the Ciris can be said to assert that of a well-cultivated garden. Whereas for Choerilus his belatedness in poetry – as far as we can judge by the extant fragments – was a disadvantage, the Ciris requires preexistent poetic tradition as the medium in which it comes into being. The other motif, that of the servant of the Muses, undergoes a similar transformation as we recognise a Lucretian subtext behind the Ciris passage (4.1177‒ 1184): at lacrimans exclusus amator limina saepe floribus et sertis operit postisque superbos unguit amaracino et foribus miser oscula figit; quem si iam admissum uenientem offenderit aura una modo, causas abeundi quaerat honestas et meditata diu cadat alte sumpta querella stultitiaque ibi se damnet, tribuisse quod illi plus uideat quam mortali concedere par est.

Strikingly enough, on a literal reading, the narrator of the Ciris is thus presented as, not a servant, but a lover – in fact, an exclusus amator – of the Muses. Con Campbell () .  It is not entirely certain that the Cypria fragment named only five flowers, since the transmitted καλλιρρόου may be a corruption for καὶ λειρίου rather than καλλιπνόου, cf. recently Parlato (). But even so, λείριον would in a sense only duplicate ἴον, both being types of lily (note that in the hymn the earlier passage has ἴα and the later λείρια instead).  Note that calt(h)a is extremely rare in classical Latin poetry, with only one context in Virgil (Ecl. .) and two in Ovid (Fast. . and Pont. ..). Remarkably, the former Ovidian context belongs to a catalogue of flowers picked by Persephone and her girl-friends.

54

1 The proem (lines 1‒100)

versely, the Lucretian unhappy lover is himself in a sense a poet, too (meditata diu cadat alte sumpta querella, cf. cantus meditanti mittere). This allusion may appear merely frivolous, but at the same time it thematises a concept that is fundamental to the Ciris: the idea of poetic inspiration as love. We already have traced 11 blandum liceat deponere amorem through Lucr. 1.924‒925 incussit suauem mi in pectus amorem | Musarum to 1.19 incutiens blandum per pectora amorem. In a similar way, I suggest, when the Ciris poet speaks of himself as only a novice to serious poetry (42‒43 sed quoniam ad tantas nunc primum nascimur artes, | nunc primum teneros firmamus robore neruos), he is alluding to Lucretius’ detailed physiological account of the onset of puberty (4.1037‒1038 sollicitatur id in nobis, quod diximus ante, | semen, adulta aetas cum primum roborat artus, 1043 in loca conueniens neruorum certa ⁹⁵). In view of the anti-Platonic colouring of Lucretius’ treatment of love in the finale of the fourth book,⁹⁶ it is interesting to note that the analogy between eros as the instinct for procreation and eros as the stimulus for intellectual creativity is, in fact, quite Platonic (Symp. 207‒209). I would conclude by suggesting that this Platonic background may be relevant for the Ciris as well. For Plato, love is what triggers the process of recollection as it directs one’s attention to the real essence (beauty) of one’s object of love. In later chapters I shall argue that the Ciris poet conceptualises poetic composition along similar lines, as an activity requiring a profound knowledge of poetic tradition that is akin to the philosopher’s knowledge of the phenomenal world rooted in knowledge of ontological reality. Whilst reserving a fuller discussion until later (in particular, 3§3, 3§8, 5§7, 6§6), I would suggest that we can already observe this poetics at work in the passage we are discussing. If my argument that the flower catalogue in the Ciris is modelled on more than one analogous context in earlier poetry is correct, it can be said that what the poet has imitated is not individual catalogues, but the very ‘idea’ of such a catalogue induced from several particular examples. Indeed, as we have seen, without a generalised knowledge of the motif of the flower catalogue the Ciris poet would have been unable to distinguish between the five invariable flower species and the sixth that can vary.

 Thus, teneros firmamus robore neruos is apparently a double entendre.  See Brown () ‒.

2 Beginning from the beginning (lines 101‒205) 2.1 Apollonius’ Argonautica begins with a word denoting ‘beginning’ (ἀρχόμενος) and moreover – like the Odyssey before it (ἄνδρα) and the Aeneid after it (arma) – with the first letter of the alphabet.¹ The ἀρχή Apollonius is talking about has several connotations, both explicit and implicit. On the most literal level, it refers to the act of composition, or rather performance, which traditionally opens with the invocation of a deity, in this case Apollo who is thus introduced as Apollonius’ patron.² At the same time, Apollo plays a crucial role in the beginning of the story itself, for it is an oracle given by him that starts the chain of events leading to the expedition.³ Yet, however necessary a condition this oracle may be, it hardly is a sufficient one, and at the beginning of the Argonautica Apollonius names, or at least hints at, further causes.⁴ Apollo’s oracle merely warned Pelias against a man with one sandal; it may be a pure coincidence that this man turns out to be Jason, who has lost a sandal when crossing the river on his way to Iolcus. As seems likely, it was on this occasion that Jason helped Hera, disguised as an old woman, over the river, and that may actually be the reason why he could not save his sandal. In any event, Jason’s service to Hera is one of the two main causes, if not for the expedition itself, then at least for its success. The other cause is Hera’s hatred for Pelias, whom she wanted to punish by bringing Medea to Greece; as the reason for this Apollonius names the omission of Hera from the sacrifices Pelias was performing when he saw Jason come without a sandal, but the literary tradition knows a more serious motive: the killing of Sidero, the stepmother of Pelias’ mother, at Hera’s altar.⁵ In view of this tradition, suppressed but not explicitly denied by Apollonius, it is Pelias’ crime and Hera’s desire to avenge it that are the ultimate causes of the Argonautic expedition. Apollonius’ interest in the remote causes, ἀρχαί,⁶ of the story he tells is confirmed further by the allusion to the famous beginning of Euripides’ Medea, which, conversely, presents the Argonautic expedition as the remote cause of the tragedy’s plot. By casting the Argonautica as a ‘prequel’ of the  On the beginning of the Argonautica see in general Levin () ‒ and Clare () ‒.  Albis () ‒.  Clare () ‒.  Levin () ‒.  Hunter () .  Cf. the invocation to the Muse at the beginning of the Aeneid (.): Musa, mihi causas memora…

56

2 Beginning from the beginning (lines 101‒205)

Medea,⁷ this allusion establishes its teleological cause and thus makes it in a sense an aetiological poem, a poem whose beginning is in its end. In fact, the aetiological dimension of the Argonautica has other aspects as well, although at first they may not be present as evidently.⁸ Indeed, the position of the Argo’s name as the last word in the very first period of the poem seems to imply some sort of etymological connection with the word it begins with, ἀρχόμενος.⁹ Likewise in the Ciris the opening of the narrative part, after a lengthy proem, explores various aspects of beginning. As with the Argonautica, particular attention is given to the circumstances that have caused the events constituting the poem’s plot: as we shall see, the Ciris poet plays, just as Apollonius did, with the impossibility of naming a single moment that could be considered the absolute beginning of the story. While discussing the proem, I have already pointed out that the allusion to Catullus 65, a preface to the translation of Callimachus’ aetiological elegy, emphasises the aetiological dimension of the Ciris; in the beginning of the narrative part we shall find further details that prefigure the poem’s end, thus announcing its teleological raison d’être. We shall also see references to the beginnings of other poems, which is another way of underlining the theme of beginning.

2.2 The narrative part of the Ciris opens in an ecphrastic mode, with a verb in the present tense (sunt); this is, in itself, a deeply traditional beginning, but it conceals, rather than announcing, what I have presented as the central theme of the first narrative section of the poem. Indeed, the first nine verses (101‒109) are merely a topographical and historical excursus introducing the reader to the scene of the forthcoming story, the city of Megara, an excursus that does not appear to have any obvious connection with the story itself. It is only in the next six verses (110‒115) that we hear of the actual historical circumstances behind the narrative, the Cretan invasion of Megaris prompted by the hospitality of the Megarian king Nisus who had provided refuge to Polyidus, a personal enemy of Minos. The next thirteen verses (116‒128) reveal a further, crucial piece of in-

 Schmakeit () ‒.  Most importantly, the Argonautica can be read as an aetiology of the Graeco-Egyptian state of the Ptolemies, see Mori ().  On the Argo as the first ship (though on the explicit level the idea is absent in Apollonius), see recently Bär (a); cf. further Gärtner () on the motif of seafaring as the origin of evil.

2.2

57

formation: Nisus had a magical purple lock on his head, and a prophecy said the city would be impregnable as long as this purple lock remained intact. Finally, from the further four verses (129‒132) we learn that, despite his confidence, Megara was sacked, because his daughter Scylla had fallen in love with Minos. As can be seen even on the surface level, each of the three last passages state a specific circumstance – Nisus’ hospitality towards Polyidus, the prophecy about the purple lock, and Scylla’s love for Minos – that in effect causes the fall of Megara, whereas the first passage mentions what is (apparently) the most fundamental cause, the very fact of building Megara. But beneath the surface the picture is even more complex. After situating Megara on the map (101‒104), the Ciris proceeds to give an account of the city’s construction (105‒109): stat Megara, Alcathoi quondam murata labore, Alcathoi Phoebique; deus namque affuit illi, unde etiam citharae uoces imitatus acutas saepe lapis recrepat Cyllenia murmura pulsus et ueterem sonitu Phoebi testatur amorem.

This account is pointedly elliptical: the narrator does not specify the way Apollo helped Alcathous, and so the reader is at liberty to surmise – wrongly – from the much better known story of the construction of Thebes that Apollo acted in the same way as Amphion, moving blocks with the sound of his lyre. But in reality Apollo helped with his sheer physical strength, and the marvellous stone obtained its musical qualities simply because Apollo, while he was working, put his lyre on it. All this we learn from an anonymous epigram of unknown date (APl 279), which arguably is a model of the Ciris passage: τόν με λίθον μέμνησο τὸν ἠχήεντα παρέρπων Νισαίην· ὅτε γὰρ τύρσιν ἐτειχοδόμει ᾿Aλκάθοος, τότε Φοῖβος ἐπωμαδὸν ἦρε δομαῖον, λᾶι¹⁰ Λυκωρείην ἐνθέμενος κιθάρην. ἔνθεν ἐγὼ λυραοιδός· ὑποκρούσας δέ με λεπτῇ χερμάδι τοῦ κόμπου μαρτυρίην κόμισαι.

The dependence of the Ciris on this epigram is supported by the fact that the Ovidian variant of the story (Met. 8.14‒18) parallels precisely those details in the epi-

 λᾶι scripsi : λᾶα [sc. δομαῖον] codd. I plan to discuss the text and interpretation of this epigram in greater detail elsewhere.

58

2 Beginning from the beginning (lines 101‒205)

gram that are omitted by the Ciris. ¹¹ But what connection does the musical stone have with the Scylla story? Why is it mentioned at all? In Ovid’s version the stone seems to provide the reason for Scylla’s falling in love with Minos: Scylla was used to climbing the walls and throwing pebbles at it before the war, and so when the war began, she would still climb the walls and, instead of playing with the stone, study the besiegers (Met. 8.19‒20). Yet in the Ciris nothing of the sort is ever suggested; moreover, the lyre stone is in no explicit way chronologically related to the main story: the stone is only said to originate in an indefinite past (105 quondam) and still to exist at present (107 etiam). The explanation seems to be that in antiquity the relative chronology of the siege of Megara by the Cretans and of the construction of the city walls by Alcathous (with Apollo’s assistance) was a matter of controversy; it is in response to this uncertainty, I suggest, that the author of the Ciris is so indirect. Alcathous was believed to have married the daughter of, and subsequently succeeded, a certain Megareus, king of Megara, but the identity of this Megareus was disputed. One tradition seems to identify him as the son of Deucalion and thus the first (and eponymous) king of Megara, while another takes him to be the successor of Nisus. Accordingly, Alcathous would have either built the original city walls, or rebuilt them after they had been destroyed by Minos.¹² Ovid obviously takes sides with the first version (hence he refers to Megara as 8.7‒8 urbs Alcathoi, quam Nisus habet), whereas the epigram is at least ambivalent, but in fact can be taken to presuppose the other chronology (note that the city is referred to as Νισαίη), for perhaps the stone should not be expected to have survived the destruction of the city by Minos. The Ciris preserves the ambiguity of the epigram and makes a point of it: whereas a casual reader would probably surmise that the order in which the construction of the walls (105‒109) and the Cretan siege (110‒111) are mentioned represents their actual relative chronology, a more sophisticated reader might observe that, if the rebuilding of the walls after the fall of Megara is actually meant,¹³ the whole plot of the Ciris turns into an aeti-

 In the Ciris, unde etiam citharae uoces imitatus and sonitu … testatur reproduce ἔνθεν ἐγὼ λυραοιδός and τοῦ κόμπου μαρτυρίην; in Ovid, turris (Met. .), deposuisse lyram (), exiguo … lapillo () and resonantia saxa (, cf.  uocalibus … muris) correspond to τύρσιν, ἐνθέμενος κιθάρην (note that the Ciris uses cithara in the context where the epigram speaks of a λύρη, whereas Ovid does exactly the opposite when he renders κιθάρη with lyra), λεπτῇ | χερμάδι and λίθον … ἠχήεντα respectively.  See Seeliger ().  Cf. Bartels () : “Im Proömium hingegen sagt der Erzähler ausdrücklich, daß Megara von Minos vollkommen zerstört wurde (), und später prophezeit er Nisus, daß kaum hohe Türme übrigbleiben werden ( f.).”

2.3

59

ology for the musical stone.¹⁴ Such an aetiological reading is supported by the fact that the phrase uoces imitatus (107) comes from the first line of a passage in Lucretius (5.1379‒1435) that gives an account of the invention of music (uoces imitarier: in the same metrical position).¹⁵ In other words, the Ciris could thus be read as in a sense an allegory of the origin of music and, by extension, poetry.¹⁶

2.3 The allusive potential of the first nine verses is far from exhausted by the reference to the epigram on the Megarian lyre stone. Even more prominent is the allusion to the opening of the Hecale, although its implications are largely lost to us because of the fragmentary state of Callimachus’ poem:¹⁷ ᾿Aκταίη τις ἔναιεν Ἐρεχθέος ἔν ποτε γουνῷ πέμπελος 〈ἄκληρός τε〉, τίον δέ ἑ πάντες ὁδῖται ἦρα φιλοξενίης· ἔχε γὰρ τέγος ἀκλήιστον.

As others have suggested, the very first word of the Hecale may be evoked in the Ciris by the first word of the second verse of its narrative part (101‒105):¹⁸ sunt Pandioniis uicinae sedibus urbes Actaeos inter colles et candida Thesei purpureis late ridentia litora conchis, quarum non ulli fama concedere digna stat Megara, Alcathoi quondam murata labore.

 Cf. O’Hara’s (: ‒) discussion of chronological inconsistencies in Catullus .  Note also Lucretius’ use of unde etiam within this passage (.). One could further speculate that (citharae) uoces … acutas may be alluding to h.Merc.  (συρίγγων) ἐνοπὴν … τηλόθ’ ἀκουστήν (in the context of the invention of the pan-pipes by Hermes, but also of his giving the lyre,  κίθαριν, to Apollo, cf.  Cyllenia murmura). This (acutas ~ ἀκουστήν) would be one of a few examples of ‘translation with paronomasia’ in the Ciris.  It is perhaps not irrelevant that Lucretius’ account of the invention of music in Book  is an important subtext for Virgil’s eclogues, see e. g. Breed ().  This reconstruction of the beginning of the Hecale was tentatively proposed by Hollis (); it joins what previously has been known as fragments  and  and adds a new one (πέμπελος).  Hollis () . On ᾿Aκταῖος, instead of the ordinary ᾿Aττικός, as a signpost for Callimachean allusion, cf. Hollis () : “The adjective ᾿Aκταῖος […] is not attested before Call.: its great popularity in subsequent Greek and Roman verse […] must be primarily due to this line of the Hecale.”

60

2 Beginning from the beginning (lines 101‒205)

In fact, virtually every word of the Hecale’s first verse finds a parallel in the quoted passage of the Ciris: τις ἔναιεν can be compared with sunt, Ἐρεχθέος ἔν … γουνῷ with Actaeos inter colles, and ποτε with quondam. Furthermore, the fame of Megara (fama) with its walls built by a deity (murata) parallels Hecale’s good reputation (τίον) for her hospitable roof (τέγος). The reference to Thesei litora, arguably implying Troezen rather than (as is generally assumed) Megaris,¹⁹ may also point to the Hecale, where Troezen plays a prominent role as Theseus’ childhood place. A further echo of the Hecale’s opening can be found in the next passage of the Ciris that introduces the specific historical circumstance of the Cretan invasion (110‒115): hanc urbem, ante alios qui tum florebat in armis, fecerat infestam populator remige Minos, hospitio quod se Nisi Polyidos auito Carpathium fugiens et flumina Caeratea texerat. hunc bello repetens Gortynius heros Attica Cretaea sternebat rura sagitta.

As is noted by Hollis, “Hecale’s hospitality […] is perhaps the most important theme of the whole poem”²⁰ (note ἦρα φιλοξενίης); in the Ciris, Nisus’ hospitality (hospitio) towards Polyidus is a sine qua non of the war and hence of the entire story (compare also τέγος and texerat). This allusion to the beginning of the Hecale at the beginning of the main narrative part of the Ciris is evidently of a programmatic nature as it invites a systematic comparison between the two poems; even though it is only fragments that we have of the Hecale, we still shall be able to notice further echoes of it in the Ciris ²¹ (cf. below 3§1, 3§9, 3§10). The significance of the motif of hospitality is further enhanced, I suggest, by a possible Theocritean allusion (12.27‒29):²²

 Cf. Waszink () : “Quamobrem, dubitanter quidem, sumpserimus poetam hic aliam urbem spectasse quae non minus artis vinculis cum Theseo coniuncta est, et quidem T r o e z e n a , quae a Statio (Theb. ,) Theseia Troezen vocatur […]. Et illic enim litus arenosum est […]; praeterea inter Athenas et Troezena complures urbes extant, ut sunt Epidaurus et Methone, quae cum Megaris conferri possunt.”  Hollis () .  The Callimachean associations of the passage may be further enhanced by a possible echo of Aet. fr.  καὶ νήσων ἐπέτεινε βαρὺν ζυγὸν αὐχένι Μίνως (referring to Minos’ thalassocracy) in populator remige Minos: both contexts seem in some way to be related to the death of Minos’ son Androgeos. The parallel is noted by Harder () .  Verbal correspondences are striking: ante alios can be compared with περίαλλα [v.l. περὶ ἄλλων], florebat (in armis) with ἀριστεύοντες (ἐρετμοῖς), remige with ἐρετμοῖς, hospitio with ξεῖνον, Nisi with Νισαῖοι, and Attica with ᾿Aττικόν.

2.4

61

Νισαῖοι Μεγαρῆες, ἀριστεύοντες ἐρετμοῖς, ὄλβιοι οἰκείοιτε, τὸν ᾿Aττικὸν ὡς περίαλλα ξεῖνον ἐτιμήσασθε, Διοκλέα τὸν φιλόπαιδα.

Theocritus here makes use of a rather obscure, but locally important, story of a certain Diocles who “fled from Attica to Megara and subsequently met his death in battle while defending his ἐρώμενος”²³ and later was commemorated by games in his honour. It is easy to observe that Diocles provides a paradigm for both Polyidus, who likewise fled to Megara, and Nisus, who likewise died defending his friend. Thus it can be suspected that the reference to Nisus’ ancestral hospitium may in fact imply not the (otherwise unattested) bonds of ξενία between his and Polyidus’ families, as is generally assumed, but rather the traditional hospitality of the Megarians, as exemplified in the story of Diocles, and of their rulers par excellence. Theocritus adduces this story as the αἴτιον of a kissing contest which was held at Diocles’ tomb every spring, and that would further enhance the aetiological dimension of the Ciris.

2.4 We have seen that in the Argonautica Apollo’s oracle about Pelias (1.5 τοίην γὰρ Πελίης φάτιν ἔκλυεν) is apparently the most immediate reason for the expedition and certainly the first we hear of. The prophecy concerning Nisus’ purple lock (119 responsum quoniam satis est meminisse deorum) is similarly of crucial impact on the consequent narrative (120‒125): nam capite ab summo regis, mirabile dictu, candida caesaries florebat tempore utroque, at roseus medio surgebat uertice crinis: cuius quam seruata diu natura fuisset, tam patriam incolumem Nisi regnumque futurum concordes stabili firmarunt numine Parcae.

Both kings misinterpret what they have learnt about their future, and for that reason, instead of avoiding the danger, they themselves bring about their own downfall: Pelias, by sending Jason to Colchis; Nisus, by being over-confident about his city’s safety and thus refusing to make peace with Minos (cf. 380). Yet prophecy can be the cause of a narrative in another sense, too, for an oracle not only starts a chain of events, but in a way already contains, in an encrypted  Gow () .

62

2 Beginning from the beginning (lines 101‒205)

form, everything that is to happen. A poem can thus be seen as both an account of an actual story and an interpretation of a prophecy. This second, poetological, meaning is explicated by the allusion to the Parcae’s prophetic song at the wedding of Peleus and Thetis in Catullus 64 (306‒309):²⁴ ueridicos Parcae coeperunt edere cantus. his corpus tremulum complectens undique uestis candida purpurea talos incinxerat ora, at roseae niueo residebant uertice uittae.

The length of the thread they spin, as well as of Achilles’ life, is determined by the length of their song; similarly, Nisus will live as long as his purple hair – like the Parcae’s thread – is not cut, and as long as he lives, the poem cannot end.²⁵ From another perspective, it may not be irrelevant that Catullus’ somewhat comic depiction of the Parcae’s feebleness is apparently modelled on the figure of Phineus (Arg. 2.197‒201),²⁶ who is, as Apollo’s prophet, an alter ego of Apollonius within the narrative.²⁷

2.5 So far we have discussed three causes which contribute to bringing about the story of the fall of Megara. First, the construction of the city walls, either before or after the Cretan invasion (in the former case it would be, so to speak, a material cause, in the latter a teleological one). Second, Nisus’ hospitality towards Polyidus which prompts Minos to besiege Megara. Third, the prophecy about Nisus’ purple lock which prevents him from making peace with Minos. All these circumstances constitute necessary conditions for the central event of the story to happen, but not the immediate cause. Looking at it from another

 In addition to the highlighted verbal parallels, Ciris  residebat,  ritu,  tereti, dente, and  custodia can be compared with Catul. . residebant,  rite,  tereti,  dens, and  custodibant respectively. The allusion is signalled by the fact that Nisus’ lock is referred to as roseus instead of purpureus, to match the colour of the Parcae’s uittae (cf. below §).  Note that both the Parcae’s thread and Nisus’ lock are being bitten: Catul. . decerpens aequabat semper opus dens, Ciris ‒ quem fibula ritu | Cecropiae tereti nectebat dente cicadae.  Papanghelis () ‒. In turn, Phineus can be argued to be a parody of Posidippus’ self-presentation in his sphragis (SH ), which seems further to confirm Phineus’ poetological significance.  Cf. Albis () ‒.

2.5

63

angle, we have seen how two important characters, Minos and Nisus, are brought to the scene and their motives revealed. And now, as the protagonist, Scylla, makes her entrance, we finally learn of the most direct cause of the fall of Megara (129‒132): nec uero haec urbis custodia uana fuisset (nec fuerat), ni Scylla nouo correpta furore, Scylla, patris miseri patriaeque inuenta sepulcrum, o nimium cupidis Minoa inhiasset ocellis.

This cause is Scylla’s love for the leader of the besiegers. But the quoted passage, being exceptionally rich in intertextual resonances, says more than that. The most important subtext comes from a fragment of Euphorion whose subject is the story of the Taphian king Pterelaus and his daughter Comaetho, a story virtually identical to that of Nisus and Scylla, the main difference being that Pterelaus’ magic lock was not purple, but golden (fr. 26.ii.14‒19 Lightfoot):²⁸ οὐ γάρ κεν νήσοισιν Ἐχινάσιν ἐσκίμψαντο οἳ [ ]νεον Κεφάλοιο καὶ ᾿Aμφιτρύωνος αμο[ ἐκ [δὲ τ]ρίχα χρυσέην κόρσης ὤλοψε Κομ[αιθὼ πα[τρ]ὸς ἑοῦ – ὡς δή ῥ’ ἄταφος τάφος εἷο πέλοιτο – εἰ μὴ ληιδίῃσι γύας ἐτάμοντο βόεσσι Τηλεβόαι διὰ πόντον ἀπ’ ᾿Aρσίνοιο μο[λόντες.

First of all, we should note the syntactical similarity of the two passages: both are unreal conditional periods, with both the protases (ni, εἰ μή) and apodoses (nec uero, οὐ γάρ) being negative. Given our (and, as I argue, the Ciris poet’s) interest in ‘causality’, we may note that a conditional period is a way of expressing a causal relationship of two events. In the Ciris Scylla’s love for Minos is thus presented as the cause of the fall of Megara; in Euphorion, however, the treason committed by a girl who has fallen in love with her city’s enemy (16‒17) is itself made a consequence of a crime perpetrated by her own compatriots (18‒19) and, by implication, of the enemy’s decision to avenge it (14‒15). By alluding to Euphorion, the Ciris appears to intimate the impossibility of specifying one single cause of an event. On the literal level, however, it is still Scylla who is made ultimately responsible for what happens, and this fact invites us to focus our attention on her figure. The most striking verbal parallel with the fragment of Euphorion is found in

 The parallel is noted in Latte ()  n. , Spanoudakis () , and Gatti () ‒.

64

2 Beginning from the beginning (lines 101‒205)

the description of Scylla as patris miseri patriaeque inuenta sepulcrum, which apparently paraphrases ὡς δή ῥ’ ἄταφος τάφος εἷο πέλοιτο.²⁹ The exact sense of the ὡς clause in Euphorion is not clear (what is its syntactical function? what is its grammatical subject? what is the referent of εἷο?), and arguably it is deliberately ambiguous. Apparently, the Ciris poet took it to be a consecutive clause,³⁰ with Comaetho as its subject and Pterelaus as the referent of εἷο: “Comaetho cut off the golden lock from her father’s head so that, by doing so, she became his grave which was no (proper) grave.”³¹ An elegiac fragment, tentatively attributed to Parthenius, points in the same direction, as it calls Comaetho παῖς Ταφίη (SH 964.11), which could mean not only ‘the Taphian girl’, but also ‘the grave-bringing girl.’³² (In a similar way, the beginning of Prop. 4.4 – Tarpeium scelus et Tarpeiae turpe sepulcrum – suggests an etymological connection between Tarpeia and turpis.) The ominous connotations of this allusion to Euphorion are further deepened by an Apollonian subtext speaking of Scylla the monster, which is signposted by the reduplicatio of Scylla in Ciris 130‒131 (Arg. 4.827‒829):³³ ἠὲ παρὰ Σκύλλης στυγερὸν κευθμῶνα νέεσθαι, Σκύλλης Αὐσονίης ὀλοόφρονος, ἣν τέκε Φόρκῳ νυκτιπόλος Ἑκάτη, τήν τε κλείουσι Κράταιιν.

 Euphorion’s oxymoron may also be behind Catul. . funera Cecropiae nec funera portarentur. Moreover, the immediate Catullan context (.‒ magnanimum ad Minoa uenit sedesque superbas. | hunc simul ac cupido conspexit lumine uirgo) is likewise connected with the present Ciris passage ( o nimium cupidis Minoa inhiasset ocellis). This can apparently be interpreted as a ‘window’ allusion through Catullus to Euphorion.  Groningen ()  views it as such, but he does not explain the optative. The general rule is that in a consecutive clause the optative may be used by attraction to the optative in the ruling clause; could the unreal modality of the aorist with a potential particle (ἐσκίμψαντο with κεν) account for the optative (πέλοιτο)? The more widespread interpretation takes the clause as a parenthetic curse by the author directed at Comaetho, cf. e. g. Lightfoot () : “may her burial be no burial!”  Cf. Latte ()  n. : “deshalb ziehe ich vor, den Satz final zu fassen: damit sie für ihren Vater ein Grab würde, das keines ist. So hat der Cirisdichter verstanden.”  Cf. Spanoudakis () , who also suggests that the fragment may belong to Parthenius. Remarkably, in this elegy the story of Pterelaus and Comaetho is juxtaposed with that of Nisus and Scylla in a sort of catalogue.  Suggested by Wills () : “The recombination of the two Scyllas was a poetic favourite, so the reference is not impeded by the fact that Apollonius’ Σκύλλη is the sea peril rather than the daughter of Nisus. In fact, the Scylla of the Ciris turns out to be just as ruinous (patris … sepulcrum) as the fabled monster (ὀλοόφρονος). The passage from Apollonius may have had further appeal as a rare mention of the monstrous Scylla’s father, since the relationship of father and daughter is at the heart of the Latin poem.”

2.5

65

This accumulation of intertextual references provides us with a rather sinister portrait of the protagonist, a portrait which suggests that Scylla’s responsibility for the ensuing tragedy can be viewed not only as an objective cause, but also in moral terms. The shift of emphasis from external objective circumstances towards internal psychological motives, which here is only adumbrated, will soon be fully accomplished. Before we turn to the next large section, we may consider two further subtexts that are at play in the last quoted passage of the Ciris. First, the expression describing Scylla’s insatiable gazing at Minos, cupidis … inhiasset ocellis, evokes the famous picture of Mars lying in the embraces of Venus from the opening lines of Lucretius’ poem, which is the only comparable attestation of inhiare in an amatory context: pascit amore auidos inhians in te, dea, uisus (1.36). This is a programmatic allusion, however incidental the parallel may look at first sight. To begin with, the immediate context in Lucretius features a striking thematic correspondence with the situation of the Ciris. Lucretius prays to Venus that her love may placate Mars and so bring peace to Rome (1.37‒40); Scylla, too, will attempt to stop war by love, with disastrous consequences. Yet this allusion may be more important as a reference to the beginning of Lucretius’ poem, which announces a systematic relationship between the two poems as wholes. Love will indeed be the central theme of the Ciris, and Lucretius’ treatment of love will recurrently be referred to. Now, though this time it may be a mere coincidence, in poetry the exact collocation nimium cupid- is only attested (apart from the Ciris) in Lucretius, in the account of the Athenian plague closing the last book of his poem (6.1239 uitai nimium cupidos mortisque timentis). But if this is not a coincidence, the Ciris would be alluding to both the beginning and the end of Lucretius’ poem in a single line, while also encompassing the two major principles – birth and death – of the Epicurean universe. Second, Scylla’s falling in love with Minos at first sight is modelled on Catullus’ account of Ariadne’s first encounter with Theseus (64.85‒86, 91‒93): magnanimum ad Minoa uenit [sc. Theseus] sedesque superbas. hunc simul ac cupido conspexit lumine uirgo … non prius ex illo flagrantia declinauit lumina, quam cuncto concepit corpore flammam funditus atque imis exarsit tota medullis.

The motif of the ‘greedy eyes’ (cupidis … ocellis, cupido … lumine) is only the most obvious shared detail. This Catullan subtext is, however, important not so much for what it has in common with the Ciris, but rather for its differences. Whereas Catullus proceeds to describe Ariadne’s psychological condition imme-

66

2 Beginning from the beginning (lines 101‒205)

diately after he mentions the fact of her seeing Theseus, the Ciris has a lengthy digression, explaining why Scylla was to fall in love with Minos, before the allusion to Catullus is resumed (163‒164): quae simul ut uenis hausit sitientibus ignem | et ualidum penitus concepit in ossa furorem… As we have seen, Scylla’s love for Minos is presented as the most direct cause of the fall of Megara, and now, in turn, Scylla’s love will be accounted for – let us repeat, in contrast to Catullus, where Ariadne’s love for Theseus does not require any external motivation – as only a consequence of some other causes. It is this digression that we now turn to.

2.6 Once more, instead of a single reason, we are given a number of alternatives. First, and perhaps not unexpectedly, we learn that Scylla’s falling in love was brought about by Amor (133‒138); immediately after that, however, we are told that Amor was not acting on his own accord, but had been prompted by Juno (139). The next passage explains why Juno wanted Scylla punished: when still a child, Scylla partook in a festival in Juno’s honour and, by accident, desecrated the goddess’ shrine (139‒153; as we remember, Pelias is punished by Hera because he neglected her when sacrificing to other gods). Yet, again, the direct cause of Juno’s anger with Scylla seems to have been not so much the breaking of a taboo itself as the girl’s false oath that she had not committed it (154‒155). Finally, one further motive for Juno’s hostility towards Scylla is revealed: the goddess’ notorious jealousy (156‒157). The exact meaning of these two lines is not quite clear (on the surface level, it would seem that they merely explain why Juno chose Amor, instead of Jupiter, to avenge Scylla’s perjury), but they certainly invite speculation that the real cause of Juno’s actions was nothing other than jealousy, whereas Scylla’s sacrilege would have been a mere pretext. Such a suspicion is confirmed by a Catullan allusion earlier in this section (137‒139): ille [sc. Amor] etiam diuos omnes – sed dicere magnum est. idem tum tristis acuebat paruulus iras Iunonis magnae…

The Catullan subtext alluded to is 68.138‒141:³⁴

 As is noted by Schwabe () . The standard interpretation of Ciris  is “he even (subdues) all the gods – but it is too large a theme to expound” (omnes is a conjecture for homines, to avoid the anticlimax: ‘gods, men’). As I argue in Kayachev  on the basis of the Catullan par-

2.6

67

saepe etiam Iuno, maxima caelicolum, coniugis in culpa flagrantem contudit iram, noscens omniuoli plurima furta Iouis. atqui nec diuis homines componier aequum est…

We should particularly note that whereas in the Ciris the nature of Juno’s anger (tristis … iras) remains unspecified, the anger of the Catullan Juno (flagrantem … iram) is certainly that of jealousy.³⁵ In the overview given above we have omitted yet another ‘cause’, which at first sight looks like a decorative vignette, but in fact has far-reaching intertextual resonances. As we are emphatically told, it is only because Scylla had loosened her dress (which otherwise would have hindered her movements) that she had the physical possibility of running into some forbidden part of Juno’s shrine, thus effecting its desecration (149‒155): …cum lapsa e manibus fugit pila quoque ea lapsa est procurrit uirgo. quod uti ne prodita ludo aureolam³⁶ gracili soluisses corpore pallam. omnia quae retinere gradum cursusque morari possent, o tecum uellem tunc, semper, haberes. non, numquam uiolata manu sacraria diuae iurando, infelix, nequiquam periurasses.

It has been noted long ago that the reference to Scylla’s golden cloak is reminiscent of the way Catullus speaks of the Golden Fleece (64.5 auratam … pellem),³⁷ but the parallel has always been treated as an inert borrowing at best. As is also well-known, Catullus in his turn alludes to the prologue of Ennius’ Medea Exul (which closely follows that of Euripides’ Medea):³⁸ utinam ne in nemore Pelio securibus caesa accidisset abiegna ad terram trabes, neue inde nauis inchoandi exordium cepisset, quae nunc nominatur nomine

allel, the line should be explained and emended in a different way: ille etiam diuis homines… sed dicere magnum est, “he even (equals) men to gods – but it is arrogant to say that.”  Just as in a Virgilian context arguably alluding to the Ciris: Georg. .‒ hoc quondam monstro horribilis exercuit iras | Inachiae Iuno pestem meditata iuuencae. Cf. further Callimachus, H. .‒ ζηλήμονος Ἥρης | χωομένης and Moschus, Eur.  χόλον ζηλήμονος Ἥρης.  As Doeblin ()  and Thomas ()  point out, Greene’s auratam may be preferable to Housman’s aureolam (emending aurea iam).  Bellinger () .  See e. g. Thomas () ‒ and Gärtner () ‒.

68

2 Beginning from the beginning (lines 101‒205)

Argo, quia Argiui in ea delecti uiri uecti petebant pellem inauratam arietis Colchis, imperio regis Peliae, per dolum. nam numquam era errans mea domo efferret pedem Medea animo aegro amore saeuo saucia.

Now we can further observe that the Ciris, in the manner of a ‘window’ reference, imitates the grammatical structure of the Ennian model, a sort of unreal conditional period (utinam ne / quod uti ne [εἴθ’ ὤφελ〈ε〉 μή] in the ‘protasis’ and numquam [οὐ γὰρ ἄν] in the ‘apodosis’).³⁹ This is a very important allusion, however unobvious it may look. To begin with, Ennius’ – or rather Euripides’ – concerns with the ‘causes’ of Medea’s tragedy are evidently much more explicit than what we see in the Ciris. It is not impossible that Herodotus, in his famous proem explaining the causes of the Persian wars, was thinking of Euripides’ prologue (1‒2 εἴθ’ ὤφελ’ ᾿Aργοῦς μὴ διαπτάσθαι σκάφος | Κόλχων ἐς αἶαν κυανέας Συμπληγάδας…)⁴⁰ when he wrote that the abduction of Medea had started the chain of offences eventually leading to the Persian invasions (1.2):⁴¹ μετὰ δὲ ταῦτα Ἕλληνας αἰτίους τῆς δευτέρης ἀδικίης γενέσθαι. καταπλώσαντας γὰρ μακρῇ νηὶ ἐς Αἶάν τε τὴν Κολχίδα καὶ ἐπὶ Φᾶσιν ποταμόν, ἐνθεῦτεν … ἁρπάσαι τοῦ βασιλέος τὴν θυγατέρα Μηδείην. At any rate, Apollonius may have connected the two passages, for in an allusion to the Euripidean context he inserts a reference to Phasis, mentioned only in Herodotus: Κολχίδα μὲν δὴ γαῖαν ἱκάνομεν ἠδὲ ῥέεθρα | Φάσιδος (2.1277‒1278).⁴² Shortly we shall see that on the intertextual level Scylla’s playfulness, as the cause of her future misfortunes, is likewise imbued with associations of a cosmological scale. At the moment, however, we can note that the Ennian allusion also points in the opposite direction as well. In the Latin rhetorical tradition of the first century BC Ennius’ prologue was a standard negative example of an excessively detailed exposition narrating events that do not have a direct bearing on the main subject.⁴³ As is pointed out by the author of the Rhetoric to Herennius (a text arguably alluded to in the proem of the Ciris, cf. above 1§4), it would have sufficed to say (2.34): utinam ne era errans mea domo efferret pedem | Medea

 Cf. further Trimble () ‒, on the reception of the Ennian utinam ne motif in Catullus.  On the ‘historiosophical’ implications of Euripides’ prologue, see Bär (b).  Pace Asheri () : “Herodotus possibly knew the story also through Euripides’ famous play.”  Note also Apollonius’ use of Κολχίδα (as in Herodotus) rather than Κόλχων (as in Euripides).  Cf. Gärtner () .

2.7

69

animo aegro amore saeuo saucia. ⁴⁴ In a similar vein, though from a philosophical rather than rhetorical perspective, Cicero criticises Ennius’ prologue for the confusion of mere antecedents and efficient causes of an event. On the one hand, he sarcastically suggests that Ennius could have made the chain of antecedents much more impressive, by beginning with utinam ne in Pelio nata ulla umquam esset arbor or even utinam ne esset mons ullus Pelius (Fat. 35). On the other, he points out (or so it would seem, the text is not quite clear) that Ennius omits the actual cause of Medea’s falling in love with Jason (presumably, the very fact of their meeting one another). It may not be irrelevant that Cicero illustrates the difference between antecedents and real causes with an example that creates a further interesting link with the Ciris: the fact of having come to the Campus Martius is not, Cicero says, the cause that makes one play with a ball (Fat. 34). As we remember, in the Ciris it is precisely Scylla’s playing with a ball that leads to the desecration of Juno’s shrine. This may be, of course, just a coincidence; but whether or not the Ciris poet was thinking of that particular passage in Cicero, he must have been fully aware of the criticism Ennius’ prologue was subjected to. The fact that at the head of the causal chain leading to Scylla’s falling in love with Minos he puts such a blatantly irrelevant situation seems to indicate this awareness.

2.7 Let us now have a closer look at the passage that gives an account of Scylla’s playing with a ball during a cult procession in Juno’s honour. This is obviously an unsuitable occupation for anyone taking part in a religious ceremony, and we are right to ask why the poet shaped the scene in exactly this way. One course of explanation is suggested by the allusion to Catullus’ image of a girl dropping a quince (149‒150 cum lapsa e manibus fugit pila quoque ea lapsa est | procurrit uirgo, cf. Catul. 65.19‒20 malum | procurrit casto uirginis e gremio). On its own, the Catullan prototype does not shed much light on our problem; but once we recognise its subtext in Callimachus’ treatment of the Acontius and Cydippe story (cf. above 1§2),⁴⁵ the parallel proves more telling. Most importantly, in both stories the dropping (actual or imagined) of either a ball or a quince marks the beginning of a love affair; in addition, both incidents take place

 Cf. also Cic. Inv. ..  For a recent discussion of Callimachus’ presence in Catullus , see Skinner () ‒.

70

2 Beginning from the beginning (lines 101‒205)

near a shrine and involve oaths.⁴⁶ We shall have to deal with this allusion in greater detail when we turn to the next section of the Ciris, where it is resumed in a more prominent way. At the moment, however, we may focus on the question why it is a ball rather than a quince that Scylla plays with. The answer lies, at least in part, in the fact that a ball performs a similarly crucial role in bringing about the central love affair in Apollonius’ Argonautica, where Aphrodite bribes Eros with a ball to shoot Medea and so make her fall in love with Jason (3.129‒144). Of course, it could be objected that the two balls cause the respective love affairs in different ways, but there are further significant parallels between the two contexts. Most generally, in both the Argonautica and the Ciris the love affair is presented as a result of direct divine intervention. In both poems it is Hera / Juno who has a personal interest in making the heroine fall in love with a stranger, and in both poems she engages Eros / Amor to perform the job. In both poems Eros / Amor figures as a naughty child (3.92 ἀναιδήτῳ, 93‒94 ἐμεῖο | οὐκ ὄθεται, μάλα δ’ αἰὲν ἐριδμαίνων ἀθερίζει, 97 ἐπηπείλησε χαλεφθείς, cf. 133‒134 malus ille puer, quem nec sua flectere mater | iratum potuit),⁴⁷ and in both poems he accomplishes the task by shooting an arrow into the girl’s heart (3.275‒287, cf. 158‒162). These parallels become especially telling when contrasted with Catullus’ treatment of an analogous situation: although Catullus does make Amor and Venus ultimately responsible for Ariadne’s falling in love with Minos, it is only in an authorial apostrophe that he does so (64.94‒98), whereas on the level of narrative these deities do not appear at all. The link between the two contexts can further be supported, although indirectly, by the similarities the Ciris has with yet another Apollonian passage featuring a ball, the passage where the nymphs helping the Argo through the Wandering Rocks are compared to girls playing with a ball (4.948‒952): αἱ δ’, ὥς τ’ ἠμαθόεντος ἐπισχεδὸν αἰγιαλοῖο παρθενικαὶ δίχα, κόλπον ἐπ’ ἰξύας εἱλίξασαι, σφαίρῃ ἀθύρουσιν περιηγέι· αἱ μὲν ἔπειτα ἄλλη ὑπ’ ἐξ ἄλλης δέχεται καὶ ἐς ἠέρα πέμπει ὕψι μεταχρονίην, ἡ δ’ οὔ ποτε πίλναται οὔδει…

In addition to the motif of playing with a ball, we may also note that the girls of the Apollonian simile have their clothes lifted up (κόλπον ἐπ’ ἰξύας εἱλίξασαι)⁴⁸  (fr.  

Moreover, Acontius and Cydippe meet for the first time during a festival in Apollo’s honour .). Lyne ()  only notes the similarity in very general terms. And so do the nymphs themselves, . ἀνασχόμεναι λευκοῖς ἐπὶ γούνασι πέζας.

2.8

71

much in the same way as Scylla does (144 suspensam gaudens in corpore ludere uestem). Later we shall see in greater detail why the (only) two Apollonian contexts involving a ball can be considered, to a certain extent, as a single unit; for now it will suffice to point out that the two balls are both described as περιηγής (3.138, 4.950), an attribute of the Empedoclean σφαῖρος (fr. 27.3). If we go on pursuing the literary genealogy of Scylla’s ball, we discover that both Apollonian contexts are, in their turn, related to the Homeric account of Odysseus’ meeting Nausicaa. In Homer, once again, a ball performs a prominent role in creating a relationship that has the potential, even if eventually not realised, of developing into a love affair. The story is a familiar one, so it will suffice to mention the most relevant detail: Odysseus is woken up by the noise Nausicaa’s companions make when she drops the ball into the river (6.115‒117). It may be difficult to prove decisively that the Ciris poet, when he made Scylla drop her ball, was thinking not only of Acontius and Cydippe, and Jason and Medea, but also of Odysseus and Nausicaa; yet the assumption is at least a natural one. Thus an interim conclusion can be formulated: Scylla’s playing with a ball, which is emphatically spoken of as (almost) the ultimate cause of her future passion for Minos, is not just a decorative vignette, as it might seem at first sight, but a motif that performs an important ‘structuralist’ function as it evokes the beginnings of three other classic love stories, in Callimachus, Apollonius, and Homer.

2.8 The motif of the ball, however, also has a meaning of its own, in addition to being a pointer to similar situations in other narratives. To describe it in the most general terms, the ball is a universal cosmic symbol, and appropriately Scylla’s ball, as we shall see, manifests certain cosmological connotations. Two striking Lucretian reminiscences concentrated in a short passage that speaks of Scylla’s playfulness invite the reader to look for a deeper, philosophical meaning behind the text, even though on their own they may not provide all necessary clues (142‒143): dum sacris operata deae lasciuit et extra procedit longe matrum comitumque cateruam.

One is an allusion to the famous laudes Epicuri at the very beginning of Lucretius’ poem (1.72‒74):

72

2 Beginning from the beginning (lines 101‒205)

ergo uiuida uis animi peruicit et extra processit longe flammantia moenia mundi atque omne immensum peragrauit mente animoque…

Two points of contact between Scylla and Epicurus can be established. First, they both break certain norms of religious nature: Scylla disturbs the order of a cult ceremony, Epicurus attacks religion as such. To this parallel we shall return later. Second, and more to the point, their disrespect of religion is in both cases counterbalanced by an interest in something else: Scylla’s attention is fully absorbed by the ball she is playing with, Epicurus is as profoundly devoted to the study of the universe. The other reminiscence comes from a context in Lucretius criticising an allegorical interpretation of the Cybele cult (2.628 Matrem comitumque cateruam) that takes it as a symbolic image of the Earth (2.600‒ 603). This parallel, too, deserves a more detailed discussion, but we must postpone it until later. If these two Lucretian allusions do indeed suggest that Scylla’s ball is, at least in a sense, a cosmic image, why is it so? The answer will become clear if we look once again at the literary predecessors of this ball. The obvious point to start from is the ball Aphrodite bribes Eros with, whose cosmological associations have been widely recognised.⁴⁹ Most relevant to the present argument, the σφαίρη promised to Eros has been connected with the Empedoclean σφαῖρος, which represents the state of the universe under the influence of Love. The connection is commonly accepted (even if with certain qualifications), so there is no need to rehearse the case. Instead, I shall argue that the (only) three other balls mentioned in either Apollonius or Homer likewise have Empedoclean connotations. Let us consider, first, the second Apollonian ball. As noted above, like Eros’ ball, this one, too, is characterised by the Empedoclean epithet περιηγής. Furthermore, the whole episode of the Argo’s passage through the Wandering Rocks can be argued to constitute an allegorical representation of the universe. We may begin by pointing out that all the four physical elements are involved in the creation of the obstacle the Argo has to pass through (4.924‒929): ἄλλοθι δὲ Πλαγκταὶ μεγάλῳ ὑπὸ κύματι πέτραι ῥόχθεον, ᾗχι πάροιθεν ἀπέπτυεν αἰθομένη φλὸξ ἄκρων ἐκ σκοπέλων πυριθαλπέος ὑψόθι πέτρης, καπνῷ δ’ ἀχλυόεις αἰθὴρ πέλεν οὐδέ κεν αὐγὰς ἔδρακες ἠελίοιο. τότ’ αὖ, λήξαντος ἀπ’ ἔργων Ἡφαίστου, θερμὴν ἔτι κήκιε πόντος ἀυτμήν.

 See Pendergraft (), Nelis () ‒, Stückelberger () , Kyriakou () , Campbell () ‒, Gillies ().

2.8

73

Earth is here represented by πέτραι, water by πόντος, air by αἰθήρ, and fire by φλόξ.⁵⁰ Perhaps even more to the point is the fact that this passage is specifically alluding to an important Empedoclean fragment, likewise listing the four elements (fr. 115.9‒12): αἰθέριον μὲν γάρ σφε μένος πόντονδε διώκει, πόντος δ’ ἐς χθονὸς οὖδας ἀπέπτυσε, γαῖα δ’ ἐς αὐγὰς ἠελίου φαέθοντος, ὁ δ’ αἰθέρος ἔμβαλε δίναις· ἄλλος δ’ ἐξ ἄλλου δέχεται, στυγέουσι δὲ πάντες.

We may particularly note the enjambement αὐγὰς | ἠελίου (‐οιο) shared by the two contexts: whereas the same phrase (with αὐγή in different cases) is relatively frequent within a single line since Homer (where it has seven contexts: Il. 8.480, 16.188, Od. 2.181, 6.98, 11.498, 11.619, 15.349, cf. 12.176), in enjambement it only occurs once in Homer (Il. 17.371‒372; but cf. perhaps 22.134) and does not reappear, except in Empedocles, until Apollonius. In Apollonius it has four occurrences, all in enjambement, and in at least one more context (1.647‒648) it is demonstrably alluding, as we shall have occasion to see (6§6), to the same Empedoclean fragment. Finally, the action of the girls playing with a ball, with whom the nymphs helping the Argo are compared, is described in nearly the same way (4.951 ἄλλη ὑπ’ ἐξ ἄλλης δέχεται καὶ ἐς ἠέρα πέμπει) as that of the four elements which in turn receive and expel the fallen δαίμων (ἄλλος δ’ ἐξ ἄλλου δέχεται, στυγέουσι δὲ πάντες). Turning now to Homer, we must acknowledge that, of course, the ball Nausicaa and her maids play with cannot be said to have Empedoclean connotations in exactly the same way as the two balls in Apollonius. Yet, although naturally Homer could not allude to Empedocles, it is well known that Hellenistic allegorists would read Homer through Empedoclean lenses, and it can be argued that the scene of Nausicaa’s playing with a ball was indeed interpreted in such a way. The scholia are not very helpful here, but Eustathius does provide some relevant information. To begin with, when discussing (apropos Nausicaa’s ball) the linguistic and cultural history of the ball in antiquity, Eustathius mentions, though in a casual way, the Empedoclean σφαῖρος as a ‘paronym’ of σφαίρη.⁵¹ Eusta It can be argued that the opening of Ovid’s Metamorphoses, which describes the primordial chaos, is alluding, inter alia, to this Apollonian context (especially .‒): utque erat et tellus illic et pontus et aer, | sic erat instabilis tellus, innabilis unda, | lucis egens aer. Compare instabilis tellus with Πλαγκταὶ … πέτραι, pontus with πόντος, unda with κύματι, lucis egens aer with ἀχλυόεις αἰθήρ. The main difference is the absence of fire in the Ovidian passage, which is introduced later (.) as deus et melior … natura.  Eust. .: παρὰ τὴν σφαῖραν, ᾗ παρώνυμον ὁ Ἐμπεδόκλειος σφαῖρος.

74

2 Beginning from the beginning (lines 101‒205)

thius’ reasons for doing so are far from clear, and his comment can hardly be taken as evidence that Nausicaa’s ball was connected with the σφαῖρος before him, but at least it shows that the connection is natural. More to the point is that Eustathius interprets the nymphs accompanying Artemis, with whom Nausicaa and her maids playing with a ball are compared (Od. 6.102‒109), as an allegorical representation of the forces of nature.⁵² We have already pointed out the connection of this Homeric context with the one in Apollonius where Eros is bribed with a ball; now we can observe that it is also imitated in the other Apollonian context. As is easy to see, here Apollonius reproduces chiastically the structure of the Homeric simile: in Homer girls playing with a ball are compared with nymphs; in Apollonius, on the contrary, it is nymphs who are compared with girls playing with a ball. Thus it would seem that Apollonius’ Empedoclean allusions in the episode of the Argo’s passage through the Wandering Rocks are prompted, at least in part, by an allegorical reading of the scene of Nausicaa and her maids playing with a ball, a reading that apparently is also reflected in Eustathius’ commentary. The second ball in Homer, not mentioned so far, is the one with which two Phaeacians play right after Demodocus’ song about Ares and Aphrodite (Od. 8.372‒376): οἱ δ’ ἐπεὶ οὖν σφαῖραν καλὴν μετὰ χερσὶν ἕλοντο, πορφυρέην, τήν σφιν Πόλυβος ποίησε δαΐφρων, τὴν ἕτερος ῥίπτασκε ποτὶ νέφεα σκιόεντα ἰδνωθεὶς ὀπίσω· ὁ δ’ ἀπὸ χθονὸς ὑψόσ’ ἀερθεὶς ῥηιδίως μεθέλεσκε, πάρος ποσὶν οὖδας ἱκέσθαι.

This passage, too, has points of contact with both Apollonian contexts involving a ball. On the one hand, the ball promised to Eros is described as a work of Adrastea in a manner reminiscent of the Phaeacians’ ball (Arg. 3.132‒133): περικαλλὲς ἄθυρμα | κεῖνο τό οἱ ποίησε φίλη τροφὸς ᾿Aδρήστεια.⁵³ On the other, the Phaeacians apparently play, not unexpectedly, the same game as Nausicaa with her maids and the girls of the Apollonian simile; it is, however, only in the first and last of the three contexts that the rule that the ball must not touch the ground is explicitly referred to: Od. 8.376 πάρος ποσὶν οὖδας ἱκέσθαι, Arg. 4.952 ἡ δ’ οὔ ποτε πίλναται οὔδει. The last detail, as well as the fact that

 Eust. . (on Od. .): κοῦραι δὲ Διὸς τουτέστιν ἀέρος, νύμφαι αἱ κατ’ ἀλληγορίαν τὴν πολλαχοῦ ῥηθεῖσαν, διὰ τὸ ἐξ ἀέρος ἐνδίδοσθαι τῇ γῇ τὸ θρεπτικὸν ἐν φυτοῖς καὶ γόνιμον, εἰς ὃ αἱ νύμφαι ἀλληγοροῦνται; . (on Od. .‒): αἰνιττόμενος ὁ ποιητὴς νύμφας εἶναι ἀλληγορικῶς τὰς περὶ γῆν φυσικὰς δυνάμεις τὰς τῶν φυτῶν αὐξητικάς.  Cf. Hunter () , Campbell () .

2.8

75

the ball is said to be thrown into the sky (ποτὶ νέφεα σκιόεντα), could easily suggest an allegorical interpretation of the game as a cosmological image, not unlike the one criticised by Lucretius in his discussion of the Cybele cult (2.602‒ 603): aeris in spatio magnam pendere docentes | tellurem neque posse in terra sistere terram. Such an allegorical reading is facilitated by the fact that the scene of the Phaeacians’ playing with a ball is preceded by the famous song of Demodocus, which was interpreted as an Empedoclean allegory:⁵⁴ a ‘cosmic’ dance would be a natural sequel to a cosmological song. This is the literary background of Scylla’s ball. As we have seen, the Ciris provides evidence that its author was aware of the cosmological – and more specifically, Empedoclean⁵⁵ – associations that the motif of the ball had within the epic tradition. But to what end are these associations used in the Ciris? One answer is obvious: they convey a universal dimension, which in antiquity was considered a necessary generic feature of epic, to the story told in the poem. In particular, they create a link between love as a human emotion, which is the driving force of the narrative, and love as a fundamental cosmic principle. Furthermore, from a certain perspective the causal chain leading to the central event of the poem – Scylla’s falling in love with Minos – turns out to be traced, as it were, ab origine mundi. However, it would be naïve to take these associations at face value as a sign that the poetic world of the Ciris harmoniously comprises both human and cosmic levels. As we remember, the allusion to the prologue of Ennius’ Medea at the same time both invites reading the story of the Ciris within the framework of universal history and, by evoking contemporary criticism the prologue was subjected to from the rhetorical (and philosophical) perspective, cautions against such a reading. Similar ambiguity is present in the way the Ciris employs the motif of the ball. Whilst on the one hand exploiting the motif’s allegorical potential, the Ciris simultaneously undermines, or at least qualifies, it by the allusions to the laudes Epicuri and Magna Mater passages in Lucretius. As has been demonstrated by others, the latter context is a programmatic discussion of allegory in which Lucretius denies it any independent epistemological value.⁵⁶ While admitting that the allegorist may present truth in his interpretation of an (alleged) allegory, Lucretius nevertheless contends that this truth can only be imposed on,

 See e. g. Nelis () ‒.  It should be noted, of course, that Empedocles’ σφαῖρος has a predecessor in Parmenides’ comparison of Being to a σφαίρη (fr. . εὐκύκλου σφαίρης ἐναλίγκιον ὄγκῳ). Thus perhaps it would be more correct to speak of Empedoclean and Parmenidean connotations in connection with the motif of the ball.  Jope (), Gale () ‒.

76

2 Beginning from the beginning (lines 101‒205)

not inferred from, a text that is interpreted in such a way. It is thus not irrelevant that, by means of two very close Lucretian reminiscences (142‒143 lasciuit et extra | procedit longe matrum comitumque cateruam, cf. above), Scylla is presented, as it were, abandoning Cybele’s cult procession, which is an allegorical image of the Earth, and, instead, directing her attention – after Epicurus’ manner – to the universe as it really is from a scientific point of view. Yet, paradoxically, on the literal level the object of Scylla’s attention is a ball, which stands for the universe (or the Earth) only by allegory. This paradox is especially remarkable because, as we have seen, in the epic tradition the ball is associated with the Earth precisely by means of an allegorical interpretation that is of more or less the same kind as the one rejected by Lucretius. We shall observe in the Ciris the same tension – let us say, between an Empedoclean and a Lucretian tradition – again and again, and its character will gradually become clearer. It may, however, be useful to attempt a provisional formulation. What I refer to as an Empedoclean tradition within the history of ancient epic is the direction that was developed – in fact, originated – by Apollonius in his Argonautica. As is well known, according to the expectations of ancient readers formed by the Homeric poems, an epos was supposed to provide a comprehensive worldview.⁵⁷ Apollonius’ unprecedented insight was that the way to rival Homer’s universality must lie, not so much through the creation of an alternative worldview based on the advances of philosophy and science, as through the reassessment of this comprehensiveness in literary terms, by incorporating into the epic tradition all branches of writing – including, of course, philosophy and science. It would seem that viewing Homer and Empedocles as part of one and the same tradition (or, to put it in a different way, reading Homer through Empedoclean lenses) was one of the cardinal points of Apollonius’ programme, for Empedocles thus created a bridge between heroic poetry and philosophical / scientific prose – which, however, does not necessarily imply that Apollonius accepted Empedocles’ theories.⁵⁸ Nevertheless, it seems possible to speak, at least in a sense, of a fundamental compatibility between the Empedoclean and Homeric worldviews, the most important point of contact, at least for our present interests, being that both poets shared the belief in the immortality of the soul. For both of them this belief is connected, though in different ways, with their attitudes to memory, which also means, from Apollonius’ perspective, with their conceptions of poetry. For Empedocles the connection is obvious: the

 See e. g. P. Hardie () ‒.  For example, Kyriakou () argues that Apollonius reads Empedocles through an Aristotelian lens.

2.9

77

wisdom Empedocles reveals in his poem has been accumulated in his previous incarnations, and he can only acquire it because he retains a memory of that experience. For Homer the connection is not as straightforward, but there exists a certain identity between the poetic afterlife (which is the essence of heroic epic, the sought-after κλέος ἄφθιτον) and the afterlife in Hades. To give just an example (an illustration rather than argument), the Homeric underworld is mainly inhabited by heroes who have also achieved a prominent place in the epic tradition. Whether or not Apollonius personally believed in the immortality of the soul, in the universe of the Argonautica souls do not perish after death (we shall return to this issue later, cf. below 6§6). Now, Lucretius openly challenged this belief, as well as the whole poetological system built upon it – by attacking Ennius, among other things. In a sense, it might be said that Lucretius restored the pre-Apollonian situation when the traditional epic worldview had become outdated and needed to be reassessed. However, Lucretius did it in such a way, partly because he worked from inside the epic tradition, that after him it would be impossible simply to repeat Apollonius’ achievement. Whereas Apollonius could make, so to speak, the Empedoclean voice – and, by extension, the voice of philosophy and science in general – sing in unison with the traditional epic one, the Lucretian voice would not merge with it: this is why the Virgilian epic universe is notoriously polyphonous. The Ciris appears to be doing the same thing as the Aeneid, but on a smaller scale. Its author realised that Lucretius’ place within the epic tradition was analogous to Empedocles’, but he also saw their incompatibility: whereas the Empedoclean voice may give support to the traditional epic world-view, the Lucretian voice will constantly undermine it.

2.9 The ball Eros is bribed with has yet another model that in certain ways is relevant to the Ciris. As has been noted by others (though without elaboration),⁵⁹ the scene of bribing Eros may be alluding to the story of Dionysus’ enticement by the Titans as told in an Orphic poem. The fragment is preserved by Clement (Protr. 2.17): ὃν [sc. Dionysus] εἰσέτι παῖδα ὄντα ἐνόπλῳ κινήσει περιχορευόντων Κουρήτων, δόλῳ δὲ ὑποδύντων Τιτάνων, ἀπατήσαντες παιδαριώδεσιν ἀθύρμασιν, οὗτοι δὴ οἱ Τιτᾶνες διέσπασαν, ἔτι νηπίαχον ὄντα, ὡς ὁ τῆς τελετῆς ποιητὴς Ὀρφεύς φησιν ὁ Θρᾴκιος (fr. 34 Kern)·

 West () , Pendergraft ()  n. , Campbell () ‒, Dickie () .

78

2 Beginning from the beginning (lines 101‒205)

κῶνος καὶ ῥόμβος καὶ παίγνια καμπεσίγυια, μῆλά τε χρύσεα καλὰ παρ’ Ἑσπερίδων λιγυφώνων.

It appears that the passage introducing the verbatim quotation is actually paraphrasing its immediate context, which seems also to be reflected in Apollonius: εἰσέτι παῖδα ὄντα and ἔτι νηπίαχον ὄντα can be compared with Arg. 3.134 ἔτι νήπια κουρίζοντι (cf. 1.508 ἔτι κοῦρος, ἔτι φρεσὶ νήπια εἰδώς), ἀθύρμασιν with 3.132 ἄθυρμα, and perhaps ἀπατήσαντες with 3.152 ἦ μέν τοι δῶρόν γε παρέξομαι οὐδ’ ἀπατήσω. There is, of course, also a difference: whereas Dionysus thus becomes a victim of a crime, Eros commits, as it were, a crime himself, since he is made responsible for the death of Absyrtus (4.445‒449).⁶⁰ The point is that, although the bribing of Eros looks perfectly innocent in its immediate context, its darker consequences are already foreshadowed by the allusion to the Orphic narrative. Whether or not the Ciris poet recognised Apollonius’ Orphic subtext,⁶¹ the motif of crime, crime on a universal scale, is certainly at play in the Ciris, too. At a later point we shall see how Apollonius’ account of the killing of Absyrtus is reflected in a subsequent section of the Ciris (4§6). At present a parallel can be suggested with another narrative related to the killing of Dionysus: the account of Pentheus’ dismemberment in Theocritus 26. It may be difficult to establish decisive arguments for a direct connection with the Ciris, but it seems clear that the Theocritean piece lies behind Catullus’ representation of the Dionysian procession in 64.251‒264 (also influenced by Lucretius’ account of the Magna Mater cult), which means that it would have been available to the Ciris poet, too. Curiously enough, Scylla’s chase after the ball she has dropped (149‒152 cum lapsa e manibus fugit pila quoque ea lapsa est | procurrit uirgo. quod uti ne prodita ludo | aureolam gracili soluisses corpore pallam) is reminiscent of the way Semela’s sisters pursue Pentheus (26.16‒17 Πενθεὺς μὲν φεῦγεν πεφοβημένος, αἳ δ’ ἐδίωκον, | πέπλως ἐκ ζωστῆρος ἐς ἰγνύαν ἐρύσαισαι). Both the ball and Pentheus are said to ‘run away’ (fugit; φεῦγεν), in both cases the pursuers have their clothes lifted up (soluisses corpore pallam, cf. 144‒145 suspensam gau The account of the killing of Absyrtus will be treated against this Orphic subtext in greater detail at a later point (§).  Since only a short fragment is preserved, one cannot be certain. A possible point of contact is that both Dionysus and Scylla are lured by playthings out of a place where they would have been protected, the former from falling victim to a crime, the latter from committing a crime. This may look a superficial parallel, but Spenser appears to have made the connection: in the story of Britomart, which is modelled on the Ciris (see Hughes [] ‒, Burrow [] ‒), Scylla’s ball is, as it were, replaced by a mirror in which Britomart sees Artegall and goes to look for him, much in the same way as Dionysus is led to the place of slaughter by a reflection he sees in a mirror, cf. Edwards ()  n. .

2.10

79

dens in corpore ludere uestem | et tumidos agitante sinus aquilone relaxans; πέπλως … ἐρύσαισαι), and in their pursuit they profane either the shrine (154 uiolata manu sacraria diuae) or the sacred objects (26.13 σὺν δ’ ἐτάραξε ποσὶν … ὄργια Βάκχω). Finally, we may point out perhaps the most remarkable parallel for the motif of crime in the Ciris: Empedocles’ fragment 115. The relevant passages in both the Ciris and Empedocles have serious textual problems, which make it difficult to reconstruct the thread of thought in either of them precisely and to develop secure arguments for a direct connection. Nevertheless, there is a very conspicuous point of contact: in both cases the crime consists, or so it seems, of both perjury (139 periura, 155 iurando … periurasses; 115.4 ἐπίορκον ἁμαρτήσας ἐπομόσσῃ)⁶² and some sort of defilement (141 uiolauerat, 154 uiolata manu; 115.3 γυῖα †μιν†, generally emended to μιήνῃ). On its own this parallel may look feeble indeed, but it makes perfect sense if considered in the context of other, more certain, Empedoclean allusions in the Ciris. For one thing, as we shall see later on (4§3), in the Ciris the motif of animal sacrifice is likewise intertextually linked to its treatment in Empedocles (this motif is relevant to the fragment under discussion because the reason for the prohibition of animal sacrifice is precisely Empedocles’ doctrine of metempsychosis presented in it). At the moment, however, it will suffice to observe that the punishment is also similar in both cases: the Empedoclean δαίμων falls into a cycle of reincarnations, and Scylla is transformed into the ciris. But again, before we proceed to the next section of the Ciris, we should make it clear that the negative connotations of the motif of crime are at the same time undermined by the presence of a Lucretian subtext. Whereas the (possible) allusions to ‘Orpheus’, Theocritus and Empedocles we have been discussing link Scylla’s almost innocent fault to the archetypal examples of an utmost crime, the reminiscence from the laudes Epicuri associates it with Epicurus’ attack on religion, which Lucretius hails as a great exploit.

2.10 The last section of the Ciris (158‒190) we shall be dealing with in this chapter describes Scylla’s falling in love with Minos, which is the most immediate cause of what happens later in the poem. One obvious model has already

 In Scylla’s case the perjury consists in swearing to a lie, in Empedocles’ apparently in breaking an oath, cf. Wright’s () ‒ discussion.

80

2 Beginning from the beginning (lines 101‒205)

been mentioned: Catullus’ account of Ariadne’s falling in love with Theseus (Ciris 163‒164 quae simul ut uenis hausit sitientibus ignem | et ualidum penitus concepit in ossa furorem, cf. 64.86 hunc simul ac cupido conspexit lumine uirgo … 92‒93 cuncto concepit corpore flammam | funditus atque imis exarsit tota medullis). Differences are, however, more important than similarities. As pointed out above, unlike the Catullan version where Ariadne’s love is presented, at least on the level of narrative, in purely psychological terms, the version in the Ciris has Scylla’s falling in love with Minos effected by Amor’s arrows. Now we may add that neither Dido in Virgil nor Scylla in Ovid – the two most obvious objects of comparison for our Scylla in Latin poetry, in addition to Catullus’ Ariadne – are shot by Amor. Although the concept of Eros’ arrows is ubiquitous, the only close extant pre-Ovidian parallel for the detailed account of Amor’s shooting that we find in the Ciris comes from Apollonius. Amor in the Ciris takes out arrows from the quiver, and so does Apollonius’ Eros (160 depromens tela pharetra; 3.279 ἰοδόκης … ἐξέλετ’ ἰόν). They both direct their missiles into the victim’s heart, and in both cases the arrow’s effect is like fire (162‒163 uirginis in tenera defixit acumina ⁶³ mente. | quae simul ut uenis hausit sitientibus ignem; 3.286‒287 βέλος δ’ ἐνεδαίετο κούρῃ | νέρθεν ὑπὸ κραδίῃ φλογὶ εἴκελον). Both girls’ faces manifest their emotions, though in conspicuously different ways (180 nullus in ore rubor: ubi enim rubor, obstat amori; 3.297‒298 ἁπαλὰς δὲ μετετρωπᾶτο παρειὰς | ἐς χλόον, ἄλλοτ’ ἔρευθος): to this difference we shall return later. Finally, in both cases love is compared, either implicitly or explicitly, to a gadfly (184 oestro, 3.276 οἶστρος): this parallel, too, deserves a more detailed discussion. In the Ciris Scylla’s passion is metaphorically referred to as a gadfly (181‒ 184): atque ubi nulla malis reperit solacia tantis tabidulamque uidet labi per uiscera mortem, quo uocat ire dolor, subigunt quo tendere fata, fertur et horribili praeceps impellitur oestro…

In Apollonius a gadfly figures, rather differently, in a simile that describes Eros’ stealthy approach to his victim (3.275‒277). However, it evokes another simile

 The reading defixit acumina is Heinsius’ conjecture for defixerat omnia, which probably should be retained.

2.11

81

from earlier in the poem where Heracles, infuriated by the news of Hylas’ abduction, is compared to a bull stung by a gadfly (1.1261‒1265, 1269):⁶⁴ ὣς φάτο· τῷ δ’ ἀίοντι κατὰ κροτάφων ἅλις ἱδρὼς κήκιεν, ἐν δὲ κελαινὸν ὑπὸ σπλάγχνοις ζέεν αἷμα. χωόμενος δ’ ἐλάτην χαμάδις βάλεν, ἐς δὲ κέλευθον τὴν θέεν ᾗ πόδες αὐτοὶ ὑπέκφερον ἀίσσοντα. ὡς δ’ ὅτε τίς τε μύωπι τετυμμένος ἔσσυτο ταῦρος… … ἵησιν μύκημα, κακῷ βεβολημένος οἴστρῳ…

To begin with a formal argument, οἶστρος is a Homeric hapax (Od. 22.300), used by Apollonius only twice, which makes the verbal parallel, very close on its own, even more compelling (horribili … impellitur oestro, κακῷ βεβολημένος οἴστρῳ).⁶⁵ It is also important that, although Apollonius does not say so explicitly, Heracles’ violent reaction is – just as in Scylla’s case – that of frustrated love:⁶⁶ soon we shall see that the symptoms described by Apollonius (and imitated in the Ciris: labi per uiscera mortem, ὑπὸ σπλάγχνοις ζέεν αἷμα) are pointing specifically in this direction.

2.11 Above I have drawn attention to the fact that, unlike Medea who blushes and grows pale in turn (blushes being a sign of shame, pallor a sign of passion), Scylla experiences constant paleness (180 nullus in ore rubor: ubi enim rubor, obstat amori).⁶⁷ This is also what distinguishes Scylla from the girl of Catullus 65, who is said to blush when her mother notices a token sent by a lover (65.24 huic manat

 Cf. Nyberg () : “It is significant that the simile in . is repeated in . f. […] This verbal echo underlines the thematic connection between the episodes.” Cf. also Campbell () .  On the literary history of this particular hapax, see Thomas (), Ross () ‒, O’Hara () ‒, Campbell () ‒.  Pace Campbell () . In an imitation of the Apollonian passage, Theocritus makes this point clear by ascribing the cause of Heracles’ madness to Eros (.‒): ὃ δ’ ᾇ πόδες ἆγον ἐχώρει | μαινόμενος· χαλεπὸς γὰρ ἔσω θεὸς ἧπαρ ἄμυσσεν (on the relative chronology of Apollonius and Theocritus, cf. Hunter [] ‒).  Similarly, as we shall see, when rendering Arg. . αἰδοῖ ἐεργομένη, the Ciris substitutes fear for shame ( demptae subita in formidine uires).

82

2 Beginning from the beginning (lines 101‒205)

tristi conscius ore rubor),⁶⁸ as well as from her Callimachean prototype, Cydippe (Aristaenet. 1.10.44‒47): τοσοῦτον ἐξεφοινίχθη τὸ πρόσωπον, ὡς δοκεῖν ὅτι τῶν παρειῶν ἔνδον εἶχέ τινα ῥόδων λειμῶνα, καὶ τὸ ἐρύθημα τοῦτο μηδὲν τῶν χειλῶν αὐτῆς διαφέρειν. The conflict of shame and love is in fact a standard ingredient in most ancient representations of enamoured heroines.⁶⁹ Neither Dido nor the Ovidian Scylla yields to passion without a fight.⁷⁰ The former gives in (Aen. 4.91 nec famam obstare furori) only after a conversation with her sister; the latter decides to surrender (Met. 8.74‒75 altera iamdudum succensa cupidine tanto | perdere gauderet, quodcumque obstaret amori ⁷¹) in the course of a tense internal monologue. Why does the Ciris poet break the standard pattern? In general, I believe, the answer is that he wanted to produce, for reasons we shall discuss later, a pure ‘case study’ of love uncontaminated by other emotions. (The fact that Scylla’s name was etymologised as αἰσχύνη⁷² – ‘shamefulness’ or, perhaps more to the point, ‘shamelessness’ – may possibly have influenced the choice of heroine.) This change of paradigm can be illustrated by the use of Callimachus’ Acontius and Cydippe narrative as a subtext. Certain points of resemblance between Scylla and Cydippe have already been noted; but there are also details that link Scylla to Acontius. To begin with, Scylla intends to send her father’s purple lock to Minos as a sort of love token (186 furtiue … mitteret): this evidently evokes, by alluding to Catul. 65.19 missum … furtiuo munere, Acontius’ trick with the quince (cf. Aristaenet. 1.10.28 λάθρᾳ⁷³). If we look back, Acontius was taught this trick by

 Catul. . missum sponsi furtiuo munere malum,  miserae,  praeceps agitur and  conscius ore rubor can be compared with Ciris  furtiue arguto detonsum mitteret hosti,  miserae,  praeceps impellitur and  nullus in ore rubor respectively.  See Farron () ‒.  In this respect, Catullus’ Ariadne is perhaps like our Scylla, since she is not described as experiencing a conflict between shame and love; however, Catullus does not speak emphatically of Ariadne’s shamelessness, as the Ciris does.  One could speculate that altera (‘the other’) is a veiled reference to the Scylla of the Ciris who yielded to her passion at once (iamdudum).  Fulg. Myth. .: Scylla enim Graece quasi αἰσχύνη dicta est, quod nos Latine ‘confusio’ dicimus; et quid confusio nisi libido est? For the text as given here (αἰσχύνη is a correction), see Jacobs () . (Cf. also Heraclit. .: Σκύλλαν δὲ τὴν πολύμορφον ἀναίδειαν ἠλληγόρησε.) One could further suggest that, instead of confusio (which in Christian texts is indeed the standard Latin equivalent for the Greek αἰσχύνη: cf. TLL, s.v. confusio, § ), the Ciris poet may have alluded to the same etymology with scelus and cognate words, which are quite frequent in the poem (, , , , , ). The wordplay on Scylla and scelus would be similar to that on Tarpeia and turpis in Prop. ...  Aristaenetus’ λάθρᾳ can plausibly be argued to go back to Callimachus, who is likely to have alluded to the Homeric hymn to Demeter ( ῥοιῆς κόκκον ἔδωκε φαγεῖν μελιηδέα λάθρῃ, ‒

2.12

83

Eros (Call. Aet. fr. 67.1‒3 αὐτὸς Ἔρως ἐδίδαξεν ᾿Aκόντιον … τέχνην⁷⁴): Scylla too, at least in Ovid’s version, makes Amor (love) responsible for her crime (8.90 suasit amor facinus). Going back one more step, we can point out that Scylla, this time in the Ciris, and Acontius are both shot by Amor / Eros, and in particular that in both cases the exceptional character of the shot is stressed (161‒162 heu nimium certo, nimium, … missu | uirginis in tenera defixit acumina ⁷⁵ mente; Aristaenet. 1.10.18‒21 ὁ Ἔρως οὐ μετρίως ἐνέτεινε τὴν νευράν … ἀλλ᾽ ὅσον εἶχεν ἰσχύος προσελκύσας τὰ τόξα σφοδρότατα διαφῆκε τὸ βέλος).⁷⁶ By drawing a parallel between Scylla and Acontius (rather than Cydippe), the Ciris poet signals that his heroine plays the role of the active lover, not of the seduced but of the seducer.

2.12 Love as disease is a commonplace in ancient literature;⁷⁷ it is no surprise that the Ciris, too, develops this topos. There are, however, specific traits that make this particular version unlike others. As we already have noted, Scylla’s lovesickness is not the physical manifestation of an internal conflict between shame and desire, but a result of her unconsummated passion alone. Looking ahead (at a context from the section we shall be discussing in the next chapter), we may point out that Carme, Scylla’s nurse, is actually able to diagnose the illness (225‒226, 237‒238): non tibi nequiquam uiridis per uiscera pallor aegroto tenuis suffundit sanguine uenas. …

 αὐτὰρ ὁ λάθρῃ | ἔμβαλέ μοι ῥοιῆς κόκκον, μελιηδέ’ ἐδωδήν): the pomegranate given by Hades to Persephone is clearly analogous to Acontius’ inscribed quince.  Harder () : “these words must refer to the trick with the apple.” On τέχνη in the sense of ‘trick’ rather than ‘skill’, cf. further Braswell (: ): “It is sometimes assumed that τέχνη has a basically neutral sense […]. However, while it is true that the word gradually takes on a broader connotation, the meaning ‘trick’, ‘ruse’ is in fact the most common in early Greek (and remains so even later).”  Retaining the transmitted defixerat omnia [sc. tela from ] would make the point even more prominent.  It is worth noting that the last motif is absent in Apollonius’ account of Eros’ shooting Medea.  In general see Funke (). For some Hellenistic examples, cf. recently Rynearson (), Faraone (), Toohey () ‒. In Latin poetry, e. g. Gill (), Merriam (), Caston (), Maltby ().

84

2 Beginning from the beginning (lines 101‒205)

ei mihi, ne furor ille tuos inuaserit artus, ille Arabae Myrrhae quondam qui cepit ocellos.

Paleness is the only directly perceptible symptom of lovesickness (furor) that is mentioned (uiridis … pallor):⁷⁸ we see the poet being consistent in this detail. These two short passages in which Carme diagnoses Scylla’s condition both imitate one and the same Apollonian context in which Chalciope observes Medea’s symptoms, but fails to interpret them correctly (Arg. 3.674‒676):⁷⁹ ὤ μοι ἐγώ, Μήδεια, τί δὴ τάδε δάκρυα λείβεις; τίπτ’ ἔπαθες; τί τοι αἰνὸν ὑπὸ φρένας ἵκετο πένθος; ἤ νύ σε θευμορίη περιδέδρομεν ἅψεα νοῦσος;

Another of Carme’s literary models, the Nurse in Euripides’ Hippolytus, although seeing the symptoms,⁸⁰ similarly learns of their cause only from Phaedra herself (347‒352). Carme’s remarkable expertise in matters of love (of which she boasts herself: 242‒243), though unprecedented in poetry, finds an intriguing parallel in a popular anecdote: the famous Hellenistic physician Erasistratus is reported to have diagnosed lovesickness in Antiochus, the son of Seleucus I, and then identified the object of Antiochus’ passion in Stratonice, his stepmother.⁸¹ The earliest extant source for the story is Plutarch’s Life of Demetrius (38.1‒12), but there seems to be no reason to doubt that, even if not necessarily authentic in every detail, it had been circulating long before Plutarch. Plutarch’s account has two interesting points of contact with the Ciris. First, both Erasistratus and Carme easily diagnose the disease, but when it comes to identifying the person who causes it, they both have difficulties: the latter has no other choice but to force a confession from Scylla; the former, since Antiochus refuses to reveal the name, recurs to systematically observing his reaction to whomever he comes into contact with. Second, much as in Erasistratus’ case the reason for the patient’s reticence is the fact that he is in love with a member of his own family, so Carme suspects – as it

 Though Carme also takes into account Scylla’s abstinence from food and sleeplessness (‒).  Cf. in particular  per uiscera and  inuaserit artus with  ὑπὸ φρένας and  περιδέδρομεν ἅψεα respectively.  Cf., among other things, Eur. Hip. ‒ τάνδ’ ἀβρωσίᾳ | στόματος ἁμέραν | Δάματρος ἀκτᾶς δέμας ἁγνὸν ἴσχειν and Ciris ‒ nam qua te causa nec dulcis pocula Bacchi | nec grauidos Cereris dicam contingere fetus (the parallel is noted by Dal Zotto [] , De Gianni [] ‒).  For a good discussion, see Rynearson () ‒.

2.12

85

turns out, wrongly – that Scylla’s passion is incestuous. Another noteworthy detail in Plutarch’s narrative, though one perhaps that may well be his own addition, is that Antiochus’ symptoms are explicitly identified as those described by Sappho in her famous poem (fr. 31 Voigt). As has been remarked in this connection, “Plutarch’s telling of the tale suggests that knowledge of the literary tradition is as essential to the diagnosis of Antiochus’ lovesickness as the physician’s medical knowledge and celebrated subtlety of perception.”⁸² It appears that in the Ciris, too, the reader is expected to diagnose Scylla’s illness with more medical precision than Carme does, by recognising allusions to the poetic tradition. As we saw earlier, the passage metaphorically referring to Scylla’s desire as a gadfly alludes to a simile in Apollonius comparing Heracles to a bull stung by a gadfly. Now I would suggest that the same Apollonian context lies behind Carme’s description of Scylla’s pathological condition. In that previous Latin passage we were able to link, in particular, 182 labi per uiscera mortem with 1.1262 κελαινὸν ὑπὸ σπλάγχνοις ζέεν αἷμα. Now we can see that the latter passage is even closer to the Greek model, with 225 per uiscera rendering (as in 182) ὑπὸ σπλάγχνοις and 226 aegroto … sanguine ⁸³ (more closely than in 182) κελαινὸν … αἷμα. Although ‘black blood’ (with either μέλαν or κελαινόν) is a standard Homeric expression, it has been suspected that “κελαινὸν here is not an epitheton ornans of αἷμα, but is pathological.”⁸⁴ It seems possible to describe Heracles’ state in more precise medical terms: the boiling of black blood is indicative, I would suggest, of an attack of ‘melancholy’ caused by an excess of black bile. This is not the place to discuss ancient medical theories, instead I propose two arguments of a literary kind. One is that Heracles was considered a typical ‘melancholiac’, that is a person with a natural superfluity of black bile which occasionally takes pathological forms: we shall consider the evidence later. The other, more directly relevant, is that in an imitation of the Apollonian passage Virgil makes the point more explicit by replacing ‘black blood’ with ‘black bile’ as the physical cause of Heracles’ frenzy (Aen. 8.219‒220): hic uero Alcidae furiis exarserat atro | felle dolor. ⁸⁵ Thus it would appear that when Carme speaks of aegroto sanguine, what is implied is blood infected with (excessive) black bile. Now, as has been amply documented by Toohey, ancient literary representations

 Rynearson () .  Though it should be noted that both MSS and editors differ in making aegrotus an epithet of either sanguine (aegroto) or uenas (aegrotas).  Platt () , cf. Vian ()  n. .  The allusion is noted by Nelis () , though he does not specifically connect atro felle with κελαινὸν αἷμα.

86

2 Beginning from the beginning (lines 101‒205)

of lovesickness often match in symptoms those of melancholy,⁸⁶ and the connection is made explicitly in some late-antique medical writings.⁸⁷ As we can see, the Ciris poet appears to have been aware of the tradition assimilating lovesickness to melancholy; but the extent and significance of his engagement with it is yet to be recovered.

2.13 Let us consider another possible literary model for the description of Scylla’s lovesickness, this time from a rather unexpected source. Nicander’s discussion of the ‘arrow-poison’ (toxicum) contains a number of suggestive parallels and, as we shall see, it provides an important key for understanding the concept of love passion underlying the Ciris. We may begin with symptomatology. Nicander describes the symptoms of the ‘arrow-poison’ taken in a drink, though he acknowledges that it is the same poison as used for smearing arrows (Alex. 207‒ 248). The Ciris speaks of the effect of Amor’s arrows, which infect Scylla’s body with a pathogenic substance: 163 uenis hausit sitientibus ignem (note the metaphor of drinking), 182 labi per uiscera mortem. The most general point of contact between the two descriptions, though perhaps too general to establish a firm connection on its own, is that both victims suffer from violent mental disorder (213 κακῇ … ἄτῃ, 164 ualidum … furorem). It could be objected that other, more detailed, parallels may result from the very fact that both contexts describe the same pathological condition (madness), even if caused by different agents; but at least some of the parallels feature more than just a generic resemblance. Admittedly, unsteady step (242 ἤμασιν ἐν πολέεσσιν ἀκροσφαλὲς ἴχνος ἰήλαι, 171 multum illi incerto trepidant uestigia cursu) may indeed be a case of such a generic resemblance. Both texts, however, not only describe symptoms in a straightforward way, but also employ similes, two each, to visualise the victim’s state of frenzy: Alex. 215‒220: δηθάκι δ’ ἀχθόμενος βοάᾳ ἅ τις ἐμπελάδην φὼς ἀμφιβρότην κώδειαν ἀπὸ ξιφέεσσιν ἀμηθείς, ἢ ἅτε κερνοφόρος ζάκορος βωμίστρια Ῥείης,

 Of course, the ancient concept of melancholy, on which see Toohey (), should not be confused with the modern understanding.  Toohey (). As Toohey shows, literary depictions of lovesickness as resembling depressive melancholy are predominantly late (Theocritus  being perhaps the earliest example), whereas the standard picture since the classical period is that of lovesickness resembling manic melancholy.

2.13

87

εἰνάδι λαοφόροισιν ἐνιχρίμπτουσα κελεύθοις, μακρὸν ἐπεμβοάᾳ γλώσσῃ θρόον, οἱ δὲ τρέουσιν, Ἰδαίης ῥιγηλὸν ὅτ’ εἰσαΐωσιν ὑλαγμόν.

And Ciris 165‒168: saeua uelut gelidis Edonum Bistonis oris ictaue barbarico Cybeles antistita buxo infelix uirgo tota bacchatur in urbe, non storace Idaeo fragrantis uncta capillos…

The point of comparison is not exactly the same in the two passages: Nicander is describing the inarticulate bawling of the poisoned, the Ciris Scylla’s rabid movements. The first similes of each passage do not have much in common (in Nicander it is a head that just has been cut off but still attempts to speak, in the Ciris it is a bacchante); but the second ones do. Both Nicander and the Ciris use for illustration the image of Cybele’s ecstatic priestess participating in the cult procession.⁸⁸ On the formal level, we may in particular point out that the two most closely corresponding verses (Alex. 217 and Ciris 166) are (almost) precise metrical and prosodical equivalents. In both of them the grammatical subject is expressed by a word denoting ‘priestess’, which in both cases occupies the same position within the verse: βωμίστρια and antistita. Outside these two corresponding verses, Scylla’s raging all over her city (167 tota bacchatur in urbe) closely resembles that of Cybele’s priestess through the public highways (218 λαοφόροισιν ἐνιχρίμπτουσα κελεύθοις). The parallel is further corroborated by the fact that “unlike the Bacchante, the ministrant of Cybele is not a conventional comparison for the emotionally distracted.”⁸⁹ Indeed, the only comparable case seems to be the speculation of the Euripidean chorus that Phaedra’s frenzy may be inspired by the Corybants or Cybele (Hipp. 141‒144 ἦ σύ γ’ ἔνθεος, ὦ κούρα, | εἴτ’ ἐκ Πανὸς εἴθ’ Ἑκάτας | ἢ σεμνῶν Κορυβάντων φοι|τᾷς ἢ ματρὸς ὀρείας;). If this parallel is not a coincidence, what role does Nicander’s discussion of the ‘arrow-poison’ play as a subtext in the Ciris? To answer this question we have to consider, in turn, two subtexts that are at play in the Alexipharmaca. Of par-

 The parallel is noted by Scaliger () , but not in the modern commentaries. Lyne ()  takes antistita to imply, “with a touch of perverseness, even humour,” the emasculated priest (like ministra at Catul. .), but nothing indicates that it is not the genuinely female priestess who is meant.  Lyne () .

88

2 Beginning from the beginning (lines 101‒205)

ticular importance is the passage at the end of Nicander’s discussion where he briefly informs on the ‘titular’ use of the poison (244‒248): τῷ μὲν Γερραῖοι νομάδες χαλκήρεας αἰχμὰς οἵ τε παρ’ Εὐφρήταο ῥόον πολέοντες ἀρούρας χραίνουσιν· τὸ δὲ πολλὸν ἀναλθέα τραύματα τεύχει σάρκα μελαινομένην· πικρὸς δ’ ὑποβόσκεται Ὕδρης ἰός, σηπόμενον δὲ μύδῳ ἐκρήγνυται ἔρφος.

The most obvious subtext here is the only explicit, and therefore highly intriguing, Homeric reference to poisoning arrows, a passage where Odysseus is said to have gone to Ephyra in search of a poison, ὄφρα οἱ εἴη | ἰοὺς χρίεσθαι χαλκήρεας (Od. 1.261‒262).⁹⁰ We see Nicander borrow from the Homeric passage the adjective χαλκήρης,⁹¹ which is extremely rare after Homer. Yet perhaps even more interesting is that ἰός, meaning ‘poison’, is here made to correspond, by being placed in the same metrical position at the beginning of a verse, to ἰούς, ‘arrows’, in the Homeric passage. This intertextual wordplay is further underlined by the use of πικρός (defining ἰός), which in Homer is a standard epithet for words meaning ‘arrow’ (most often ὀιστός). A similar pun is employed in the Theriaca, where Nicander applies (at 184) the adjective ἰοδόκος, a standard Homeric epithet of the quiver meaning ‘arrow-storing’, to the poisonous fangs of the asp, reinterpreting it as ‘poison-storing’.⁹² As Overduin observes in this connection, such a reuse of a Homeric word is characteristic of the way Nicander “aligns himself with the epic tradition and its master”, as he “presents us a world in which snakes and scorpions do not attack on instinct, but go to war.”⁹³ The ‘arrow-poison’ thus turns out to be an archetypal poison, poison par excellence, as with its specific name (toxicum) it reduplicates the constituent metaphor of the Nicandrian epic universe – poison as arrow – that is implied in the generic noun ἰός. The Homeric allusion works also on another level: at Od. 2.228‒230 one of the suitors voices a concern that Telemachus may have gone to Ephyra to procure a poison he intends to administer into their wine; this is, apparently, the same substance as mentioned at Od. 1.261‒262, and thus it would parallel the double use of the ‘arrow-poison’ in Nicander.⁹⁴ As is noted by Jacques, from the iologist’s point of view the expression Ὕδρης ἰός allows of two interpretations: it may be implying either that the

    

On which see Dirlmeier (). Cf. also West ()  and Jacques () . Cf. Jacques () . See Overduin () . See Overduin () . Cf. Jacques () .

2.13

89

‘arrow-poison’ is produced from a plant that sprang from the blood of Hydra killed by Heracles (so the scholia would have it), or that it is the venom of a Hydralike creature, that is of a snake.⁹⁵ There is, however, also a purely literary dimension in this expression, as it alludes to the well-established mythological fact that Heracles had his arrows smeared with Hydra’s venom. Of particular relevance seems to be Sophocles’ Trachiniae (even if Nicander’s dependence on it may be difficult to prove decisively), where the double – or, in fact, triple – use of Hydra’s venom is at the heart of the plot. First, it kills Nessus, wounded by Heracles’ poisoned arrow (573‒574): μελάγχολος | ἔβαψεν ἰὸς θρέμμα Λερναίας ὕδρας⁹⁶ (cf. 717 ἰὸς … μέλας). In passing we may note that, like Nicander, Sophocles is here apparently playing on the homonymy of ἰός: at 574 ἰός is most naturally taken to mean ‘poison’⁹⁷ (cf. 717), whereas only ten lines earlier Heracles is said to have shot a κομήτην ἰόν (567). Second, when administered through a garment soaked with it (which is in a sense parallel to taking the ‘arrow-poison’ in a drink discussed by Nicander), Hydra’s venom inflicts a lethal disease on Heracles (767‒771): ἱδρὼς ἀνῄει χρωτί, καὶ προσπτύσσεται πλευραῖσιν ἀρτίκολλος, ὥστε τέκτονος χιτών, ἅπαν κατ’ ἄρθρον· ἦλθε δ’ ὀστέων ὀδαγμὸς ἀντίσπαστος· εἶτα φοίνιος ἐχθρᾶς ἐχίδνης ἰὸς ὣς ἐδαίνυτο.

Nicander’s striking metaphor of the poison’s eating through flesh (247‒248 πικρὸς δ’ ὑποβόσκεται Ὕδρης | ἰός) may well originate from this passage. Third, the poison is intended by Deianira, on Nessus’ misleading instructions, as a love-potion. As has been demonstrated by Faraone, the mistake by which Deianira kills her husband has parallels in the real world of the fifth-century Athens, where it was a common belief that certain poisons, administered in small doses, produced the effect of a love-potion.⁹⁸ If the ‘arrow-poison’ was indeed used for such purposes as well as for killing, or at least was thought to be,⁹⁹ it would be a  Cf. Jacques () .  I give Davies’ text (with Dobree’s nominative singular for the transmitted μελαγχόλους … ἰούς); for a discussion of this difficult passage see Davies () ‒.  But cf. Davies () : “Would ἰός mean ‘arrow’ […] or ‘poison’? In view of the accompanying epithet μελάγχολος the difference between the two renderings is hardly very great.”  Faraone (), Faraone () ‒. Cf. Carawan () ‒, Lee () ‒ .  It is worth noting that, for example, Plautus can speak of Amor’s toxicum (Cist.  uideo ego te Amoris ualde tactum toxico), even if, admittedly, it stands for poison in general rather than for the ‘arrow-poison’ specifically.

90

2 Beginning from the beginning (lines 101‒205)

sufficient reason for the Ciris poet to make an allusion to Nicander’s discussion of it. In fact, at 161 Amor’s tela are given the epithet Tirynthia, usually thought to be corrupt because “it can make no sense here”;¹⁰⁰ but it could make sense if understood as a reference to Heracles’ practice of smearing arrows with what is not only a poison but also, at least potentially, a love-potion.¹⁰¹

2.14 As already mentioned, in antiquity Heracles was considered a typical melancholiac; the famous Aristotelian ‘problem’ 30.1 explaining why most outstanding men, of whatever occupation, are characterised by a superfluity of black bile is here of particular relevance. Remarkably, it interprets Heracles’ symptoms, which in Sophocles’ account are caused by intoxication with Hydra’s venom, as the manifestation of an attack of black bile (953a): καὶ ἡ πρὸ τῆς ἀφανίσεως ἐν Οἴτῃ τῶν ἑλκῶν ἔκφυσις γενομένη τοῦτο [sc. that Heracles was a melancholiac] δηλοῖ· καὶ γὰρ τοῦτο γίνεται πολλοῖς ἀπὸ μελαίνης χολῆς. The latter version is, of course, a rationalising variant of the traditional mythological story, but the way it reinterprets its source is more subtle than is obvious at first sight. It is hardly a coincidence that at Tr. 573‒574 Sophocles refers to Hydra’s venom (or so it seems) as μελάγχολος ἰός; and it is also not irrelevant that χολή is a term not only for ‘bile’, but also for ‘venom’. Whatever connotations μελάγχολος has in Sophocles, for a retrospective reader it would be difficult not to connect it with the μέλαινα χολή of (pseudo‐)Aristotle. In a sense, then, it can be said that Hydra’s ‘black venom’ (μελάγχολος ἰός) and Heracles’ own ‘black bile’ (μέλαινα χολή) are one and the same substance; the only difference is that the former has an external source, whereas the latter comes from inside. There is also another remarkable feature shared by the two substances, apart from being pathogenous. We have already seen that Hydra’s venom is treated by Sophocles, in accordance with popular belief, as a potential love-potion. In a similar way, (pseudo‐)Aristotle makes a direct connection between black bile and libido: as he laconically observes (953b), οἱ μελαγχολικοὶ οἱ πλεῖστοι λάγνοι εἰσίν. According to (pseudo‐)Aristotle, black bile has also one further noteworthy quality, namely it is connected with intellectual excellence, among other things

 Lyne () . For instance, Goold, in Fairclough and Goold () , prints ferientia, conjectured by an anonymous friend of Schrader’s.  Perhaps it could be interpreted literally, too, for after all Heracles and Eros are half-brothers (much in the same way as Aeneas and Amor are), and it would not be impossible to invent a story in which Eros inherited Heracles’ bow.

2.14

91

in poetry (953a): διὰ τί πάντες ὅσοι περιττοὶ γεγόνασιν ἄνδρες ἢ κατὰ φιλοσοφίαν ἢ πολιτικὴν ἢ ποίησιν ἢ τέχνας φαίνονται μελαγχολικοὶ ὄντες; … ἔτι δὲ τῶν περὶ τὴν ποίησιν οἱ πλεῖστοι [sc. are subject to diseases caused by black bile]. Perhaps the most familiar expression of this idea is found in Horace who famously ridiculed it in the Ars poetica (301‒302): o ego laeuus, | qui purgor bilem. As Brink comments: “If inspired poetry involves madness, that is, a surplus of (black) bile supposed to cause the melancholia of genius, then (H[orace] suggests sarcastically) he would rather be rid of it, and be sane.”¹⁰² This brings us back to the observation made in the preceding chapter (1§9) that two allusions to Lucretius’ treatment of the subject of love at the end of Book 4 implicitly present the Ciris poet’s own poetic activity in erotic terms. As I suggested there, these two allusions establish for the first time in the poem the constituent poetological metaphor that defines its main allegorical level of meaning: the poet as lover. It seems relevant to point out that the same Lucretian context also contributes to the portrayal of Scylla in love (4.1117‒1120):¹⁰³ inde redit rabies eadem et furor ille reuisit, cum sibi quod cupiant ipsi contingere quaerunt, nec reperire malum id possunt quae machina uincat. usque adeo incerti tabescunt uolnere caeco.

Scylla’s inability to find a cure for her passion is referred to in words clearly reminiscent of Lucretius (Ciris 181‒182): atque ubi nulla malis reperit solacia tantis | tabidulamque uidet labi per uiscera mortem. ¹⁰⁴ Thus we can see the metaphor of the poet as lover receive a rationalist foundation: both psychological states, that of poetic inspiration and that of being in love, turn out to be caused by the same physiological agent, black bile. In the following chapters we shall observe how Scylla is further developed, through her amorous experience, as an allegorical poet figure. It is primarily because of this metaphorical meaning, I suggest, that love is given uncontested predominance among Scylla’s emotions.

 Brink () ‒.  Cf. also the collocation pedibus Sicyonia at both Lucr. . and Ciris .  Furthermore, the Lucretian phrase furor ille is used in Ciris  ei mihi, ne furor ille tuos inuaserit artus, a context that we have seen to be alluding to the same Apollonian passage as Ciris ‒.

3 The night episode (lines 206‒348) 3.1 The night episode (lines 206‒348) is the only dramatically developed part of the Ciris. Only here do we find two characters in conversation (Scylla herself and her nurse) – a dialogue which is the only event exactly situated both in time (at night) and in space (in the palace). The episode is conspicuously placed in the centre of the poem, being both preceded and followed by about two hundred lines. It is also the longest self-contained unit (over one hundred and forty lines), though it covers a relatively short timespan, just one night. Such a prominent position of the night episode in the Ciris is the more remarkable as in other extant literary sources for the Scylla story there is no nurse for Scylla to talk to in a ‘nurse scene’, which here constitutes the heart of the episode.¹ Lyne is probably right: “The clear signs are that our poet is building up or wholly introducing this scene into the story himself.”² In other words, the night episode promises to be the most significant and original part of the poem. In the night episode the Ciris comes closest to the conventions of traditional epos. As Arend points out in his seminal study of type-scenes in Homer, epic action is meticulously synchronised with the alternate change of day and night, normally taking a pause for sleep: “Schlaf ist Zäsur des Lebens; so auch der epischen Erzählung; es ist kein Zufall, daß die Schlafszenen vielfach mit den Rhapsodiengrenzen zusammenfallen […]. Der Schlaf ist eingespannt zwischen Sonnenuntergang und -aufgang.”³ The Ciris features an exact reversal of the standard epic pattern, for its central episode extends, not from sunrise to sunset, but from sunset to sunrise; it opens at the moment when Nisus has just fallen asleep (206‒207), and continues until the next dawn (349‒350), these being the only references to the time of day in the poem. Why does the Ciris reverse the traditional scheme? One answer is given in the text. At 351‒352, where the coming of the next day is described, the planet Venus is said to be welcome to girls as the morning star and unwelcome as the evening star: quem pauidae alternis fugitant optantque puellae: | Hesperium uitant, optant ardescere Eoum. The reason behind their ambivalent attitude to a

 Knox () ingeniously interprets Ov. Met. .‒ curarum maxima nutrix | nox as an allusive reference to a (lost) version of the Scylla story that, like the Ciris, featured a nurse; but if we assume that the Ciris predates Ovid, there will be no need to multiplicare entia.  Lyne () .  Arend () .

3.2

93

heavenly body is at the moment irrelevant, though later we shall discuss it (4§2).⁴ What is important is that this passage alludes to a fragment of Callimachus’ Hecale that speaks of a likewise ambivalent attitude to Venus by some (unidentifiable) males, who, on the contrary, like it when it appears in the evening and dislike it when it appears in the morning (fr. 113 Hollis): αὐτοὶ μὲν φιλέουσ’, αὐτοὶ δέ τε πεφρίκασιν, | ἑσπέριον φιλέουσιν, ἀτὰρ στυγέουσιν ἑῷον.⁵ By substituting girls for males, and ascribing to the former an attitude that is exactly the opposite to that of the latter, the Ciris apparently comments on the fact that its central action takes place, in an un-epic manner, at night: whereas in traditional epos main characters are men, for whom it is appropriate to act in daylight, in the Ciris the central scene is played by two women. It must be stressed, however, that this is a reversal, not rejection, of the traditional epic pattern, which still, as we shall see, provides a generic model for the episode.

3.2 The comparatively traditional character of the narrative in the night episode creates an excellent opportunity for observing how the principle of composition by theme that we discussed in the Introduction is manifested in written poetry. We may start with the introductory passage describing Scylla’s preparation for cutting her father’s purple lock, a passage that exemplifies what may be called the theme of ‘the beginning of a night action’ (206‒219): iamque adeo dulci deuinctus lumina somno Nisus erat uigilumque procul custodia primis excubias foribus studio iactabat inani, cum furtim tacito descendens Scylla cubili auribus arrectis nocturna silentia temptat et pressis tenuem singultibus aera captat. tum suspensa leuans digitis uestigia primis egreditur ferroque manus armata bidenti euolat. at demptae subita in formidine uires. caeruleas sua furta prius testatur ad umbras: nam qua se ad patrium tendebat semita limen, uestibulo in thalami paulum remoratur et alte suspicit ad celsi nictantia sidera mundi, non accepta piis promittens munera diuis.

 Cf. Lyne () ‒.  See Hollis () ‒, who argues, against Lyne, for a direct connection.

94

3 The night episode (lines 206‒348)

In the preceding chapter we already have pointed out some parallels between the conversation of Scylla and Carme, her nurse, and that of Medea and Chalciope, her half-sister (as well as wet nurse, note Arg. 3.734). As we shall see, Apollonius is indeed the main model for the night episode in the Ciris; it is important to remember, however, that the corresponding section of the Argonautica is more complex in structure. First, Medea goes to bed, sees a dream, wakes up, and has a conversation with Chalciope (3.616‒743). Then, left alone, she spends the rest of the night on deciding whether or not to help Jason (3.744‒824). Both these parts can be said to be conflated into a single sequence of action in the Ciris. The opening passages of both are, accordingly, reflected in that of the Ciris night episode quoted above. Let us first consider the passage that introduces the second scene, in which Medea is alone (3.744‒751): νὺξ μὲν ἔπειτ’ ἐπὶ γαῖαν ἄγεν κνέφας· οἱ δ’ ἐνὶ πόντῳ ναυτίλοι εἰς Ἑλίκην τε καὶ ἀστέρας Ὠρίωνος ἔδρακον ἐκ νηῶν, ὕπνοιο δὲ καί τις ὁδίτης ἤδη καὶ πυλαωρὸς ἐέλδετο, καί τινα παίδων μητέρα τεθνεώτων ἀδινὸν περὶ κῶμ’ ἐκάλυπτεν· οὐδὲ κυνῶν ὑλακὴ ἔτ’ ἀνὰ πτόλιν, οὐ θρόος ἦεν ἠχήεις· σιγὴ δὲ μελαινομένην ἔχεν ὄρφνην. ἀλλὰ μάλ’ οὐ Μήδειαν ἐπὶ γλυκερὸς λάβεν ὕπνος.

To begin with, the Ciris reproduces the basic compositional principle of this typescene: the contrast of the vigilant protagonist against the sleeping secondary characters; we shall see that this motif goes back to Homer. The difference, however, is that whereas in the Argonautica the grieving mother submitting to sleep has no function in the plot, in the Ciris Nisus’ sleeping is highly functional: it is only because he is asleep that Scylla can cut his purple lock. The same applies to other individual motifs: the quietness of night (σιγὴ δὲ μελαινομένην ἔχεν ὄρφνην), navigation by stars (ναυτίλοι … ἔδρακον), the guard struggling with sleep (πυλαωρός) are mentioned in the Argonautica only “to exert an hypnotic effect on the reader.”⁶ In the Ciris, by contrast, Scylla herself listens to the silence in order to ensure that nobody will intervene (nocturna silentia temptat), she likewise looks at the stars to ascertain that they are favourable (suspicit ad … sidera), whereas the vigilant guard (uigilum … custodia), protecting the city from the besiegers, is pointedly unable to save Nisus from his own daughter. It is hardly an accident that many motifs that remain passive in the Apollonian context can be so easily made fully functional. I would argue that this can  Campbell () .

3.2

95

best be explained by their ‘thematic’ nature, that is by the fact that they have specific functions in other realisations of the same theme of ‘the beginning of a night action’. To start with, let us consider the motif of the vigilant guard. In the Argonautica it points to one of the two contexts of the Homeric dis legomenon πυλαωρός (reused by Apollonius only once)⁷ that belongs to a ‘beginning of a night action’ type-scene (Il. 24.677‒681): ἄλλοι μέν ῥα θεοί τε καὶ ἀνέρες ἱπποκορυσταὶ εὗδον παννύχιοι, μαλακῷ δεδμημένοι ὕπνῳ· ἀλλ’ οὐχ Ἑρμείαν ἐριούνιον ὕπνος ἔμαρπτεν ὁρμαίνοντ’ ἀνὰ θυμόν, ὅπως Πρίαμον βασιλῆα νηῶν ἐκπέμψειε, λαθὼν ἱεροὺς πυλαωρούς.

We see the familiar basic pattern: all others are asleep, Hermes alone remains awake. The situation is particularly poignant here: during their journey to the Greek camp, which took place at dinner time, Hermes easily put the guard to sleep (Il. 24.443‒445) with his magic staff (Il. 24.343), whereas now on the way back, in the depth of night, he apparently chooses the less expected strategy of making Priam invisible. Apollonius’ full awareness of the ‘thematic’ potential inherent in the motif of the vigilant guard is confirmed by the allusion to the same Homeric context in his account of how Medea leaves the city unnoticed by the guard (Arg. 4.48‒49 οὐδέ τις ἔγνω | τήνγε φυλακτήρων, λάθε δέ σφεας ὁρμηθεῖσα, cf. Il. 24.681 λαθὼν ἱεροὺς πυλαωρούς, 691 οὐδέ τις ἔγνω). This is precisely the motif of the vain watchfulness of the guard that we find in the Ciris (note 208 inani). Similarly the motif of looking at the stars finds a functional application in another Apollonian passage developing the same theme, a passage that describes the onset of the next night and Jason’s preparation for performing Hecate’s rites (3.1193‒1197): Νὺξ δ’ ἵπποισιν ἔβαλλεν ἔπι ζυγά· τοὶ δὲ χαμεύνας ἔντυον ἥρωες παρὰ πείσμασιν. αὐτὰρ Ἰήσων, αὐτίκ’ ἐπεί ῥ’ Ἑλίκης εὐφεγγέες ἀστέρες Ἄρκτου ἔκλιθεν, οὐρανόθεν δὲ πανεύκηλος γένετ’ αἰθήρ, βῆ ῥ’ ἐς ἐρημαίην, κλωπήιος ἠύτε τις φώρ.

In a familiar way, Jason staying awake is contrasted with his companions who go to bed. Though it is not explicitly stated that he looks at the stars, this can be inferred from the remark that the Bear has declined (ἔκλιθεν): as Hunter points

 On Apollonius’ use of Homeric hapax and dis legomena, see especially Kyriakou ().

96

3 The night episode (lines 206‒348)

out, “its approach to the horizon could mark the middle of night.”⁸ Midnight is the time at which Jason has been instructed by Medea to perform the rites (3.1029), so one naturally concludes that Jason learns the time by observing the position of the Bear in the sky. The Ciris poet seems not merely to have intuited this potential implication of the motif of looking at the stars (that is, learning time rather than direction or something else), but in fact to have used precisely that context. At Ciris 218 the transmitted reading nutantia (sidera) has long been suspected, and often changed to Scaliger’s nictantia; yet though not regularly used in an astronomical context, nutare – if understood as ‘to decline (to the horizon)’ – would be a neat equivalent of κλίνεσθαι.⁹ At a later point we shall have occasion to consider the passage introducing Medea’s conversation with Chalciope in greater detail, but one particular motif can be discussed in connection with the preceding argument. When Medea attempts to go to Chalciope, she is said to leave her room barefoot (Arg. 3.646 νήλιπος), “because in a hurry and distracted.”¹⁰ Similarly Scylla leaves her room on tiptoe (212 suspensa leuans digitis uestigia primis, implying that she goes barefoot), though for a different reason: she is not absent-minded but, on the contrary, extremely cautious. It might appear that the Ciris poet takes an Apollonian motif and invests it with an entirely new meaning, but the fact is that this meaning is implicitly present already in the Apollonian context, as is revealed in a parallel scene. Two days later, when leaving for good, Medea is again described as going barefoot (Arg. 4.43 γυμνοῖσιν δὲ πόδεσσιν), but this time she does so for the same reason as Scylla, to avoid being noticed. According to Hunter, “this detail is one of many echoes and contrasts between the scenes.”¹¹

3.3 It is perhaps time to pause and look back at the material we have assembled. As we have seen, the introductory passage of the night episode in the Ciris can be related to several models. In addition to the two functionally analogous passages

 Hunter () .  OLD s.v. a: “To incline from the vertical, sink, nod.” Man. . nutantis summo de uertice mundi (i. e. the part of heaven to the west of the meridian) is perhaps the closest parallel. Cf. further Verg. Aen. . (= .) cadentia sidera, Prop. .. in Oceanum sidera lapsa [v.l. lassa] cadunt.  Hunter () .  Hunter () .

3.3

97

that introduce the scenes of Medea’s conversation with Chalciope and of her deciding whether or not to help Jason, the Ciris uses two further contexts that narrate the events of the two subsequent nights. How should we account for this complex relationship between a passage of the Ciris on the one hand and four interconnected contexts of Apollonius’ Argonautica on the other? One way of describing the process of composition is to suppose that the author of the Ciris worked in a mode similar to that in which we analysed the passage: first he would use one particularly suitable context as the main model, and then proceed to borrow individual details from other analogous contexts for embellishment. The other way is to imagine the poet as taking for his model not a number of separate contexts in sequence, but a set of similar contexts as, in a sense, an organic unity. These two speculative schemes need not, of course, be mutually exclusive, but rather may describe the same process from two different perspectives, one in diachrony and the other in synchrony. I would, however, suggest that the latter comes closer to the truth, at least in our case. As Aristotle wrote in Poetics 9 (1451b), φιλοσοφώτερον καὶ σπουδαιότερον ποίησις ἱστορίας ἐστίν· ἡ μὲν γὰρ ποίησις μᾶλλον τὰ καθόλου, ἡ δ’ ἱστορία τὰ καθ’ ἕκαστον λέγει. This famous remark on the difference between poetry and history, I believe, can help us better to understand the compositional principles underlying the poetic texture of the Ciris. Although the character of these universals or generalities (τὰ καθόλου) is sometimes disputed, most naturally Aristotle seems to mean “the plot as an abstract dynamic schema (rather than a sequence of events drawn from life).”¹² As is easy to observe, this ‘abstract dynamic schema’ looks very much like the concept of cognitive ‘script’, a mental mechanism responsible for organising memory “in terms of standardized, generalized episodes, or structures of expectations,”¹³ a mechanism that has been argued to underly the phenomenon of themes (or type-scenes) in oral poetry.¹⁴

 Whalley ()  n. . Cf. Heath (), Armstrong () (p. : “the relevant universals to be event-types”). The universal character of such a sequence of events consists in the fact that they follow each other according to necessity or probability rather than at random. Cf. Halliwell () : “[T]he principle of generality or typicality which Poetics  sketches means that the events of a dramatic poem should exhibit a higher level of intelligibility, particularly causal intelligibility, than is usually to be found in life. The plot […] is not to be understood as simply corresponding to reality past or present […], but as representing a heightened and notional pattern of possibility, and as therefore more accessible to rational apprehension than are the events of ordinary experience. On the argument of Poetics , it is not immediately to life that the poet must turn for his material, but to an imagined world (including that of inherited myth) in which the underlying designs of causality, so often obscured in the world as we encounter it, will be made manifest.”  Minchin () . I discuss this in greater detail in the Introduction.

98

3 The night episode (lines 206‒348)

It is not irrelevant that Aristotle conceives of the compositional process in terms of mimesis of reality, that is in a manner clearly analogous to the later concept of imitation – mimesis – of literary models.¹⁵ Aristotle, as Janko aptly puts it, “used μίμησις to express the way in which art represents reality, in that good art can extract general patterns of action from the jumbled disorder of events and represent them more clearly than history would.”¹⁶ I would suggest that the same distinction can be applied to the use of literary models: whereas one sort of poetry imitates, similarly to history, individual (con)texts as independent entities (even if several at the same time), good poetry will discern in them ‘general patterns of action’.¹⁷ This is, I believe, the internal logic of the ‘thematic’ principle of composition as manifested in written poetry: once the poet realises that his true object of imitation is “the underlying designs of causality,”¹⁸ he will turn, in order to extract these ‘designs’, to studying individual exemplars of a given theme or type-scene.¹⁹

 Cf. Melia () ‒: “The fact that in the Poetics Aristotle attributes the necessity for the primacy of plot to the necessity for plausibility of the causation of the action depicted ought not to blind us to the similarity his aesthetic judgment here shows to that of the SerboCroatian singers interviewed by Parry and Lord […] For the Serbo-Croatian singers, the story-pattern (= Aristotle’s plot) consists of the correct sequence of obligatory themes in the correct order.” Note further Melia’s (p. ) observation that Aristotle’s theory of rhetoric “provides, quite unconsciously, I think, the first systematic description of the mental structure of oral tradition.”  Cf. Russell () ‒.  Janko () .  Cf. Russell () : “The extant theoretical discussions of imitatio […] make two central points. One is that the true object of imitation is not a single author, but the good qualities abstracted from many […]. The second point, related to the first, is that the imitator must always penetrate below the superficial, verbal features of his exemplar to its spirit and significance. The analogy between these points and those made by Aristotle in his account of general poetic mimēsis is, I think, clear: in Aristotelian theory, all poetry deals in generalities (Poetics b), and requires not only verbal skill but, more importantly, understanding of character and plot.” Since Russell’s understanding of τὰ καθόλου is loose (“all poetry deals in generalities”), he defines the object of truly poetic imitation likewise loosely (“good qualities abstracted from many”, “spirit and significance”). I make a more specific point.  Halliwell () .  This is, of course, a simplistic argument, which I have developed mainly for the purpose of illustration rather than demonstration. Several qualifications are needed, some of which would require extensive discussion. First, Aristotle is mainly concerned with tragic plot, but epic plot follows basically the same principles; hence, when speaking of poetry, I mainly mean epic poetry, since the Ciris is an epos. Second, Aristotle seems to be dealing with the ‘abstract dynamic schema’ of whole plots rather than individual episodes, whereas my point of concern is ‘theme’ or ‘type-scene’: it can probably be argued that the same logical consistency is required on a

3.4

99

3.4 We already have observed how the poet of the Ciris reworks four passages from the Argonautica, all of which belong to the theme of ‘the beginning of a night action’, into one. Moreover, as we have also seen, he does not do so in a merely mechanical way, but he can be said precisely to ‘extract general patterns of action’ from his models, as he charges every single motif with its proper function, often obscured in the main source context. In fact, as I shall be arguing, this procedure is not confined to imitating the Argonautica, but also involves other texts. We may start with a somewhat problematic example, which might even seem to endanger the preceding argument. The Ciris passage we are dealing with contains allusions to two contexts in Catullus 64, neither of which, however, belongs to a ‘beginning of a night action’ type-scene. One allusion links Scylla’s quasiheroic night enterprise to Theseus and Ariadne’s joint victory over the Minotaur. Like Ariadne praying for Theseus (64.103‒104 non ingrata tamen frustra munuscula diuis | promittens), Scylla prays for her own success (219 non accepta piis promittens munera diuis), and like Theseus getting out from the Labyrinth (64.113 errabunda regens tenui uestigia filo, 114 egredientem), she is very cautious when she leaves her room (213‒214 tum suspensa leuans digitis uestigia primis | egreditur). There are two points that can be returned to an objection of this kind. On the one hand, it is possible for several themes to be at play within the same context: they need not be mutually exclusive. Here the common ground linking the two situations is that in both cases the daughter helps her father’s enemy; this thematic congruence might admittedly be viewed as a theme in the stricter sense. On the other hand, the two motifs deriving from Catullus are in fact so altered as to be incorporated within the motif structure of the main theme of ‘the beginning of a night action’. The motif of carefully directing one’s own steps (regens tenui uestigia filo) is ‘transformed’ into that of going barefoot (leuans digitis uestigia primis), which we have seen to derive from Apollonius. Later we shall have occasion to note that the motif of prayer likewise belongs to the theme of ‘the beginning of a night action’. In general, much as a theme is the sum of an indefinite number of contexts describing analogous situations, so (potentially) all occurrences of a motif, in whatever contexts, can be viewed as an organic unity.

smaller scale. Third, it is not self-evident whether ‘general patterns of action’ underlying poetic texts are the same as those underlying real-world events: if yes, why does one need to extract them by directing one’s attention to texts rather than events? if no, why are they of interest (i. e. what sort of reality do they belong to)?

100

3 The night episode (lines 206‒348)

This principle can further be illustrated by means of the second Catullan reminiscence in our passage, where the sleeping Nisus is described with the same expression (206 dulci deuinctus lumina somno) as the sleeping Ariadne when abandoned by Theseus (64.122 deuinctam lumina somno).²⁰ To start with, let us point out that the Ciris poet slightly modifies the expression, by adding the epithet dulci ²¹ and thus adjusting it to how the motif appears in the theme of ‘the beginning of a night action’ (cf. Arg. 3.751 γλυκερός). Moreover, the same epithet (‘sweet’, though expressed differently: γλυκίων μέλιτος) is also used in the passage that arguably is the source of the Catullan formulation, the beginning of Moschus’ Europa (2‒4):²² νυκτὸς ὅτε τρίτατον λάχος ἵσταται ἐγγύθι δ’ ἠώς, ὕπνος ὅτε γλυκίων μέλιτος βλεφάροισιν ἐφίζων λυσιμελὴς πεδάᾳ μαλακῷ κατὰ φάεα δεσμῷ…

Here we find the same idea of, literally, binding (πεδάᾳ … δεσμῷ ~ deuinctam) the eyes (φάεα ~ lumina: note that both use the same metaphor) as in the Latin phrase. It is to an extent irrelevant whether or not Catullus actually alluded to Moschus.²³ What is important is that the Ciris uses this Catullan formula (with the addition of dulci) to refer, apart from Apollonius, to Moschus (note 220‒221 Phoenicis filia … sensit ~ 7‒8 Φοίνικος θυγάτηρ … ὠίσατ〈ο〉: Scylla’s nurse turns out to be Europa’s sister). In turn, the opening scene of the Europa, which tells of a dream seen by the heroine, is heavily indebted to the night episode of the Argonautica;²⁴ and although the only action that takes place is in this dream, the quoted passage can itself be said to belong to the theme of ‘the beginning of a night action’. So, the formula ‘sweet sleep’ in the Ciris pointedly links the context in which it occurs with at least two contexts in Greek poetry that develop the same theme; or perhaps it should be said that the reference is made not so much to the two actual contexts as to the theme that they exemplify; or, to put it

 Add  descendens ~ . discedens (both in the same metrical position, producing a homodyne fourth foot). This link stresses that both protagonists behave in an adulterous way: Theseus abandons Ariadne, Scylla secretly leaves her bed like a married woman going to a lover.  The text of Catullus . is lacunose, and its restoration uncertain. Trappes-Lomax ()  and Trimble () ‒ have recently both decided against Lachmann’s supplement uenerit, preferring an epithet to go with somno: the former suggests securo, the latter accepts Laetus’ placido. Some recentiores have in fact dulci, but it seems unlikely, pace Trimble, so soon after dulcem in ..  Cf. Lyne () .  Cf. Trimble () .  See e. g. Campbell () ‒.

3.4

101

in yet another way, the object of imitation is two contexts qua realisations of a theme. The fact that the contexts in Apollonius and Moschus are also linked genetically is, from a certain perspective, only an accidental addition to their being linked logically by participation in the same theme. But how is the reuse of Catullus’ phrasing to be accounted for? It may be possible to argue that there is a common thematic ground between the two contexts, for indeed Nisus is betrayed by Scylla more or less in the same way as Ariadne is by Theseus. However, the similarity does not extend beyond these two motifs (sleep and betrayal), therefore we cannot speak of a theme in the stricter sense (as a sequence of motifs). By analogy, the object of imitation can be said to be that particular Catullan context qua realisation of a motif (rather than of a theme). This is, of course, an extremely schematic, perhaps even rigid, way of describing the process of poetic composition. No doubt, apart from the ‘thematic principle’, numerous other forces are also involved, in different combinations depending on the specifics of a given person, genre, and tradition. But the ‘thematic principle’ is a powerful force, and the Ciris author seems to have realised that fact and applied that force in a highly self-conscious way. And this has significant implications for how we conceive of the use of literary sources in the Ciris (and not only in the Ciris). It is, of course, important to register individual models that are demonstrably contributing to a particular context. However, it should be understood that every single intertextual link thus created (or rather reified) is not so much a new meaning added to the context – as is implied by such terms as allusion or reference; rather, it is a meaning chosen from a variety of potentialities that are objectively, though implicitly, present in any context by virtue of its participation of a theme. As a consequence, reading a poem should not be restricted to ‘solid’ parallels that can be confirmed by precise verbal correspondences, but should be open to any contexts developing the same theme as potentially relevant. Ultimately, since the theme is an atemporal reality like the Platonic form, there should in principle be no difference between thematically related contexts anterior to a given poem or passage and those postdating it; or, to put it less radically, between those known to the author and those unknown, between those the author wanted to use and those he would prefer to ignore. As I shall be arguing in one of the following chapters, the Ciris poet was aware of this undiscriminating power of the theme as well as of the dangers hidden in reading (or writing) a poem, so to speak, sub specie aeternitatis.

102

3 The night episode (lines 206‒348)

3.5 At the beginning of the chapter I made the observation that the night episode in the Ciris both comes closest to the standard epic pattern of action and at the same time reverses it by depicting a night action instead of a day one. Then I argued that the main model for the passage introducing the night episode is found in the Argonautica, which is known, of course, to be in constant play with the traditional epic conventions of the Homeric poems. Now I would suggest that beneath the night episode of the Ciris, as well as its models in Apollonius and Moschus, lies a specific Homeric subtext: the tenth book of the Iliad known as the Doloneia, which is the only example of an extended night action in either of Homer’s poems.²⁵ Many textual parallels with Iliad 10 have, in fact, been pointed out by earlier scholarship in both the night episode of the Argonautica and in the opening scene of the Europa. To focus on just the short passage of the latter quoted above, 2 νυκτὸς ὅτε τρίτατον λάχος ἵσταται ἐγγύθι δ’ ἠώς is alluding to 10.251‒253 ἐγγύθι δ’ ἠώς· | … παρῴχηκεν δὲ πλέων νὺξ | τῶν δύο μοιράων, τριτάτη δ’ ἔτι μοῖρα λέλειπται, 3 ὕπνος … βλεφάροισιν ἐφίζων to 10.26 ὕπνος ἐπὶ βλεφάροισιν ἐφίζανε, and 4 πεδάᾳ μαλακῷ κατὰ φάεα δεσμῷ, though less specifically, to 10.2 μαλακῷ δεδμημένοι ὕπνῳ.²⁶ As Hunter aptly comments on the Homeric subtext in the description of Medea’s sleeplessness: “Particularly important are echoes of the scene from Il. 10 where the cares of leadership keep the Greek general [Agamemnon] awake; the transference of such a scene from the military sphere to that of personal emotion is a characteristic technique of later amatory poetry.”²⁷ This is a valid observation, but it seems possible to give a more specific reason for the use of that particular model. Recently Dué and Ebbott have argued, contrary to the widespread view, that Iliad 10 is no less traditional than the rest of the poem and its specific features can be better explained not as result of interpolation, but by the fact that the Doloneia is an example of an epic theme that is almost completely absent elsewhere in the Iliad. ²⁸ In traditional Greek epic, as they demonstrate, war could be presented in two ways: either as regular combat, or as various kinds of guerrilla action. Most battle scenes of the Iliad clearly belong to the former theme (πόλεμος),

 By action I mean outdoor activities that do not normally (or at least regularly) take place at night.  See Bühler () ‒, Campbell () ‒.  Hunter () .  Dué and Ebbott (). Cf. Thornton () , who likewise explains the peculiarity of Iliad  by its being “the only book in the poem in which action by night is described.”

3.5

103

whereas the Doloneia is a rare specimen of the latter, which they label the theme of ambush (λόχος). The two themes, though to a certain degree structurally analogous, can be differentiated by a number of significant oppositions. One such opposition is particularly relevant to our argument: whereas πόλεμος is always conducted in daytime, λόχος normally, if not exclusively, takes place at night. Perhaps the most notable example of the theme of ambush outside the Iliad is the narration of the sack of Troy in the epic cycle.²⁹ Nevertheless, there are clear signs that the exceptional character of the Doloneia within the Iliad was noticed in antiquity.³⁰ This is apparently what attracted to it the attention of Apollonius and, later, Moschus: their innovative epics are thus shown, in a sense, to be grounded in the (seemingly) least conventional book of Homer’s heroic poem. Earlier we have seen how the passage introducing the night episode in the Ciris employs a significant number of motifs from several structurally analogous passages in Apollonius’ Argonautica. Now I would further demonstrate that many of these motifs, with their specific implications, originate in the Doloneia. In most cases it will be difficult to argue for the Ciris’ immediate dependence on Homer rather than Apollonius’ earlier use of Homer; but there will also be specific parallels that strongly suggest the direct influence of Iliad 10. Let us consider the introductory passage of the night episode once more, this time in its relation to the Doloneia. The contrast of the sleeping Nisus and the vigilant Scylla (206‒209) finds a neat parallel in the very first lines of Iliad 10 (1‒4): ἄλλοι μὲν παρὰ νηυσὶν ἀριστῆες Παναχαιῶν εὗδον παννύχιοι, μαλακῷ δεδμημένοι ὕπνῳ· ἀλλ’ οὐκ ᾿Aτρείδην ᾿Aγαμέμνονα ποιμένα λαῶν ὕπνος ἔχε γλυκερὸς πολλὰ φρεσὶν ὁρμαίνοντα.

This is, of course, the context that lies beneath the main Apollonian model of the Ciris passage (3.751 ἀλλὰ μάλ’ οὐ Μήδειαν ἐπὶ γλυκερὸς λάβεν ὕπνος). The transmitted reading in 206 deuictus (lumina somno), rather than deuinctus, is arguably closer to δεδμημένοι (subdued, rather than bound, by sleep). The idea of the eyes (lumina) as the organ with which sleep is experienced is paralleled later on in 10.26 ὕπνος ἐπὶ βλεφάροισιν ἐφίζανε and 91‒92 οὔ μοι ἐπ’ ὄμμασι νήδυμος ὕπνος | ἱζάνει. Finally, as we shall discuss in greater detail, Nisus turns out to

 Cf. Dué and Ebbott () ‒.  The T scholia even claim that the Doloneia was not originally part of the Iliad at all. Cf. Casali () ‒.

104

3 The night episode (lines 206‒348)

be, in a sense, a counterpart of the sleeping Rhesus (10.474 Ῥῆσος δ’ ἐν μέσῳ εὗδε). Next comes the motif of the vigilant guard. We have already seen how in the Ciris it is linked to two contexts in the Argonautica, one of which in turn goes back to a concise ‘night action’ scene in Iliad 24.³¹ In Iliad 10 this motif is even more prominent and compositionally important. Agamemnon’s sleeplessness is caused by the fear of a night attack by the Trojans, which, first of all, prompts him as well as other Achaean leaders to go and check whether the watch have not fallen asleep (97‒99 δεῦρ’ ἐς τοὺς φύλακας καταβείομεν, ὄφρα ἴδωμεν, | μὴ τοὶ μὲν καμάτῳ ἀδηκότες ἠδὲ καὶ ὕπνῳ | κοιμήσωνται, ἀτὰρ φυλακῆς ἐπὶ πάγχυ λάθωνται).³² When they find the watch duly wakeful (181‒182), they decide to send a spy into the enemy’s camp; as it turns out, the success of this mission is due to the failure of the Trojans’ allies to set up a watch (408, 416‒421). So, like Agamemnon and Medea, Scylla cannot sleep. Like Agamemnon (Il. 10.9‒10 ὣς πυκίν’ ἐν στήθεσσιν ἀνεστενάχιζ’ ᾿Aγαμέμνων | νειόθεν ἐκ κραδίης, 94‒95 κραδίη δέ μοι ἔξω | στηθέων ἐκθρῴσκει) and Medea (Arg. 3.755 πυκνὰ δέ οἱ κραδίη στηθέων ἔντοσθεν ἔθυιεν, 760 ὣς δὲ καὶ ἐν στήθεσσι κέαρ ἐλελίζετο),³³ she is deeply distressed (211 singultibus, cf. 345 crebrosque insani pectoris ictus). Like Agamemnon (10.21 ὀρθωθείς) and Medea (3.645 ὀρθωθεῖσα), she gets up from her bed (209 descendens Scylla cubili). The next two motifs – Scylla’s going on tiptoe (apparently barefoot) and arming herself with a ferrum bidens – allude to the conventions of a dressing / arming scene. As Dué and Ebbott point out in their discussion of the dressing and arming scenes in Iliad 10, they are unlike those taking place by day in at least two respects.³⁴ First, the clothing and armour involved are different, to suit the conditions of a night raid, which are different too. Second, it is common for those who go on a night mission to leave a piece of their own clothes or armour behind and be equipped with someone else’s. I would suggest that Scylla’s going barefoot may simultaneously be playing on both these points. On the one hand, it could be speculated that she simply forgets to put her shoes on (as is apparently the case with Medea

 On Priam’s visit to the Achaean camp as a variation on the theme of ambush, see Dué and Ebbott () ‒ and passim.  This proves a timely precaution, for Hector, too, is interested in whether the Achaean camp is guarded (Il. .‒, cf. ‒: note the formula καμάτῳ ἀδηκότες repeated from .).  As Hunter ()  comments, “the text closely reproduces the pattern of the Homeric model in the opening of Il. .”  Cf. Dué and Ebbott () ‒.

3.5

105

at Arg. 3.646, as we have seen).³⁵ On the other, and certainly more to the point, going barefoot is a strategy intended to avoid making noise (cf. Arg. 4.43),³⁶ much in the same way as Diomedes and Odysseus’ wearing leather, rather than metal, helmets (Il. 10.257‒258, 261‒262) allows them to hide in the darkness.³⁷ And obviously, Scylla’s description as ferroque manus armata bidenti (213) evokes an arming scene in quite a direct manner; it may, in fact, be alluding specifically to Diomedes’ being armed with a φάσγανον ἄμφηκες (10.256). From this point on, Scylla’s attempted ‘attack’ on Nisus begins to resemble the spying mission of Diomedes and Odysseus. As suggested earlier, the motif of looking at the stars may imply that Scylla learns the time from their position in the sky. Likewise, Odysseus urges Diomedes to make haste with their preparations when he notices that ἄστρα δὲ δὴ προβέβηκε (10.252). Another point of contact is that both Scylla and the ambushers pray and make promises to the gods (219 non accepta piis promittens munera diuis, Il. 10.277‒295); though unlike Scylla’s prayer, those of Odysseus and Diomedes are heard (10.295 τῶν δ’ ἔκλυε Παλλὰς ᾿Aθήνη). Finally, Scylla’s listening to the silence (210 auribus arrectis nocturna silentia temptat) finds a parallel of a sort in that Odysseus and Diomedes could hear, but not see, a heron sent by Athena (10.275‒276 τοὶ δ’ οὐκ ἴδον ὀφθαλμοῖσι | νύκτα δι’ ὀρφναίην, ἀλλὰ κλάγξαντος ἄκουσαν). As Dué and Ebbott comment, “that the bird is heard but not seen is also appropriate to the poetics of the night, where hearing is the predominant sense.”³⁸ So far it has not been my objective to argue that all these similarities, linking Scylla either to Agamemnon or to Odysseus and Diomedes, are deliberate allusions intended to create meaningful parallelism. What I have primarily tried to show is that quite a number of the motifs used in the introductory passage of the Ciris night episode can easily be found a place within the matrix of the epic theme of ambush. But now, going deeper into the night episode, to the point where the dialogue between Scylla and Carme takes place, I shall suggest a further group of correspondences, one of which seems to be close and specific enough to count as a real ‘quotation’. At this point Scylla assumes, as it were, the

 Similarly Scylla is wearing only a light dress ( tenui … crocota; cf. Arg. . οἰέανος), which prompts Carme to wrap her in her own cloak. This motif is paralleled in, and can arguably be traced back to, a concise night ambush episode in the Odyssey (.‒) that “includes one ambusher leaving behind his cloak and another dropping his so that he can run faster,” Dué and Ebbott () . As Hunter ()  points out, οἰέανος is “a ‘female’ variant of Homeric οἰοχίτων,” which is a hapax used to describe precisely that former ambusher (Od. .).  Agamemnon, however, does put his shoes on in a regular way (Il. ., cf. ).  Cf. Dué and Ebbott () .  Cf. Dué and Ebbott () , cf. ‒.

106

3 The night episode (lines 206‒348)

part of Dolon, whereas that of Odysseus and Diomedes is taken over by Carme. (Needless to say, Dolon’s mission is exactly analogous to that of the Greek spies.) Straight after he sets out on his way, Dolon is noticed by Odysseus (10.339 τὸν δὲ φράσατο προσιόντα); similarly Scylla, as she leaves her room, is noticed by Carme (221 surgere sensit). Both are caught (10.376‒377 τὼ δ’ ἀσθμαίνοντε κιχήτην, | χειρῶν δ’ ἁψάσθην, 223 corripit extemplo), and the interrogating begins. At Dolon’s plea to spare his life Odysseus replies that he should not be afraid, and then asks (10.384‒389): ἀλλ’ ἄγε μοι τόδε εἰπὲ καὶ ἀτρεκέως κατάλεξον· πῇ δ’ οὕτως ἐπὶ νῆας ἀπὸ στρατοῦ ἔρχεαι οἶος νύκτα δι’ ὀρφναίην, ὅτε θ’ εὕδουσι βροτοὶ ἄλλοι; ἦ τινα συλήσων νεκύων κατατεθνηώτων, ἤ σ’ Ἕκτωρ προέηκε διασκοπιᾶσθαι ἕκαστα νῆας ἐπὶ γλαφυράς, ἤ σ’ αὐτὸν θυμὸς ἀνῆκε;

Odysseus’ questions find a striking parallel in Carme’s first speech, a parallel that comes as close as possible to a definitive proof of the direct use of Iliad 10 in the Ciris (231‒236): …qua causa ad patrium solam uigilare cubile, tempore quo fessas mortalia pectora curas, quo rapidos etiam requiescunt flumina cursus? dic age nunc miserae saltem, quod saepe petenti iurabas nihil esse mihi, cur maesta parentis formosos circum uirgo remorere capillos.

Individual correspondences are quite obvious: ἀλλ’ ἄγε μοι τόδε εἰπέ ~ dic age nunc miserae, πῇ ~ qua causa, ἔρχεαι οἶος ~ solam uigilare, ὅτε ~ tempore quo, βροτοὶ ἄλλοι ~ mortalia pectora. The two passages are also structurally similar: both start with a general question and then suggest specific reasons; this double structure is stressed in both cases by naming twice the physical target of the expedition (ἐπὶ νῆας and νῆας ἐπὶ γλαφυράς, ad patrium … cubile and parentis … circum … capillos). Perhaps it will not be superfluous to call attention to the fact that the main point of resemblance between the two passages picks up the most basic leitmotif of the theme of night action, the contrast of the vigilant protagonist with the sleeping entourage (cf. εὕδουσι βροτοὶ ἄλλοι with 10.1‒2 ἄλλοι … εὗδον). In fact, this sort of question seems to be found only once in Homer outside the Doloneia (where it has two more occurrences: 10.82‒85 and 141‒142), namely in Iliad 24 (at 362‒363), which we already have seen to use motifs from the theme of ambush.

3.5

107

Both Dolon and Scylla tremble in reaction to the questioning (10.390 ὑπὸ δ’ ἔτρεμε γυῖα, 256 tremebunda), both start their replies by admitting that they are out of their minds (10.391 πολλῇσίν μ’ ἄτῃσι παρὲκ νόον ἤγαγεν Ἕκτωρ, 258 nostros … furores). Both ask why all these questions (10.432 ἀλλὰ τίη ἐμὲ ταῦτα διεξερέεσθε ἕκαστα; 258 quid tantum properas … nouisse?), but promise to tell everything (10.413 = 427 τοὶ γὰρ ἐγὼ καὶ ταῦτα μάλ’ ἀτρεκέως καταλέξω, 266 dicam equidem). Dolon begs Odysseus and Diomedes to spare his life (10.378 ζωγρεῖτ〈ε〉), but is killed, failing to perform the act of supplication (10.454‒455 ὃ μέν μιν ἔμελλε γενείου χειρὶ παχείῃ | ἁψάμενος λίσσεσθαι). Scylla, in contrast, begins by invoking the gods (273‒274) and only then asks for mercy (275 ut me, si seruare potes, nec perdere malis). Earlier I suggested that Scylla’s ferrum bidens can be compared with Diomedes’ φάσγανον ἄμφηκες. Now we may pursue the parallel a little further: it is with this sword that Diomedes decapitates Dolon (10.455‒456 ὃ δ’ αὐχένα μέσσον ἔλασσε | φασγάνῳ ἀίξας) and it is likewise ferro hoc (280) that Scylla considers killing herself (282 peperissem uulnere letum). Both Dolon and Scylla reveal their plans – unlike Scylla’s main prototype Medea, who in the conversation with Chalciope conceals her real motives: Dolon admits that he has been sent by Hector (10.391), Scylla confesses that she is acting in the interests of Minos, as it were (268‒272). Dolon also gives away information about Rhesus, who is another counterpart of Minos: both Rhesus and Minos receive detailed ecphrasis-like descriptions (10.435‒441, Ciris 268‒270), both are said to be in the middle of the enemy – either physically (10.435 ἐν δέ σφιν Ῥῆσος βασιλεύς, 474 Ῥῆσος δ’ ἐν μέσῳ εὗδε) or figuratively (264 media ex acie, mediis ex hostibus). There is also possibly one more point of contact between Rhesus and Minos: when the Greek spies eventually reach the Thracian camp, Odysseus points at Rhesus and his horses (10.477 οὗτός τοι, Διόμηδες, ἀνήρ, οὗτοι δέ τοι ἵπποι), in a manner reminiscent of how Scylla speaks about Minos (268 ille, uides… 272 ille mea, ille idem oppugnat praecordia Minos). To begin with, Scylla’s use of uides (when Minos is most certainly out of view) has been considered puzzling,³⁹ but it neatly captures the deictic force of Odysseus’ τοι (‘look you’). Furthermore, the repetition of ille in 272 can arguably be compared with that of οὗτος in 10.477 (though, strictly speaking, the latter is a polyptoton, not reduplication). The last parallel is supported by an analogous reduplication in Propertius 4.4, a poem that has demonstrable links with both the Ciris and Iliad 10:⁴⁰ ille equus, ille meos in castra reponet amores ⁴¹  Lyne () .  On the possible use of the Ciris in Prop. ., cf. Hutchinson (: ): “Among relevant models will have been treatments of Scylla, probably or possibly including those of: […] the Ciris (st cent. BC?).” For parallels with the Doloneia, see below n. .

108

3 The night episode (lines 206‒348)

(37). Whilst the Ciris seems to repeat the former οὗτος (οὗτος … ἀνήρ), Propertius picks out, in a complementary manner as it were, the latter one (οὗτοι … ἵπποι).⁴² Unlike Dolon, however, Scylla is not killed, but on the contrary joined, by Carme. It is Nisus, not Minos, who, like Rhesus, eventually falls victim to a night raid. A further confirmation of the link between the Ciris and Iliad 10 comes from the Nisus and Euryalus episode in Aeneid 9, with which the Ciris features remarkable textual parallels⁴³ and which in turn is largely modelled on the Iliad 10.⁴⁴ It seems worth specifically pointing out that the use of the Doloneia in the Ciris does not require Aeneid 9 as a model (and therefore can predate it), since, as noted earlier, a precedent for alluding to Iliad 10 in an amatory context would have been established for the Ciris by Apollonius and Moschus.

 The parallel is noted by Lyne () .  The horse Tarpeia is referring to is, of course, Tatius’; like those of Rhesus (. λευκότεροι χιόνος), it is remarkable for its hair ( cui Tatius dextras collocat ipse iubas). There is a spring near Tatius’ camp from which a war horse is said to drink ( bellicus ex illo fonte bibebat equus: it is, of course, a collective singular, but one naturally thinks of Tatius’ horse in the first place); about Rhesus’ horses there was a prophecy saying that once they had drunk from the Scamander, they would become invincible (the story is not Homeric, but Virgil knows it, Aen. .‒: ardentisque auertit equos in castra prius quam | pabula gustassent Troiae Xanthumque bibissent). Like Rhesus (. τεύχεα δὲ χρύσεια πελώρια, θαῦμα ἰδέσθαι), Tatius has exceptionally beautiful armour ( obstipuit regis facie et regalibus armis, cf.  formosa oculis arma Sabina meis). There are also other parallels. Agamemnon is deeply distressed as he looks at the Trojan campfires (. πυρὰ πολλά, τὰ καίετο Ἰλιόθι πρό), and so is Tarpeia when she thinks about those in the Sabine camp ( ignes castrorum). The Trojans do not have a regular watch (. οὔ τις κεκριμένη ῥύεται στρατὸν οὐδὲ φυλάσσει), though there is some military activity in the camp (. αὐλῶν συρίγγων τ’ ἐνοπήν); the Romans neglect every precaution altogether (‒ Romulus excubias decreuit in otia solui | atque intermissa castra silere tuba). Like Dolon, Tarpeia gives over vital information to the enemy, but instead of being rewarded, is killed. Like Odysseus (.‒), Tarpeia tells the time of night by the stars (‒ et iam quarta canit uenturam bucina lucem, | ipsaque in Oceanum sidera lassa cadunt). Diomedes appears to the sleeping Rhesus like a bad dream (.‒ κακὸν γὰρ ὄναρ κεφαλῆφιν ἐπέστη | τὴν νύκτ’ Οἰνείδαο πάις), Tarpeia calls Tatius to come to her as a good dream ( fac uenias oculis umbra benigna meis). The motif of the contrast of the vigilant protagonist with the sleeping entourage (Il. .‒, though more relevantly .‒) is also there (‒ omnia praebebant somnos: sed Iuppiter unus | decreuit poenis inuigilare tuis). Finally, the reference to Vesta as Iliacae felix tutela fauillae () evokes the events of the Trojan war in a general way.  See Gorman ().  See recently Casali (), Dué and Ebbott () ‒.

3.6

109

3.6 The preceding discussion has mainly focused on the passage introducing the night episode and on its subtexts. As already pointed out, the night episode in Argonautica 3, which I claim to be the main model of that in the Ciris, consists of two parts: one narrating the conversation between Medea and Chalciope and the other giving an account of Medea’s actions when she is left on her own. The discussion of individual subtexts began with the passage opening this second part of the night episode, in which the wakeful Medea is contrasted with the quiet of night (3.744‒751). But I promised to return to the first part, which is the prototype for the conversation between Scylla and Carme in the Ciris night episode. Unlike Scylla, Medea is asleep at the beginning of the night episode and has a dream, from which she wakes up in fear and agitation (3.616‒644). Next she gets up and tries to go to Chalciope (3.645‒651): ἦ ῥα, καὶ ὀρθωθεῖσα θύρας ὤιξε δόμοιο νήλιπος οἰέανος· καὶ δὴ λελίητο νέεσθαι αὐτοκασιγνήτην δὲ καὶ ἕρκεος οὐδὸν ἄμειψε. δὴν δὲ καταυτόθι μίμνεν ἐνὶ προδόμῳ θαλάμοιο αἰδοῖ ἐεργομένη· μετὰ δ’ ἐτράπετ’ αὖτις ὀπίσσω στρεφθεῖσ’· ἐκ δὲ πάλιν κίεν ἔνδοθεν, ἄψ τ’ ἀλέεινεν εἴσω· τηΰσιοι δὲ πόδες φέρον ἔνθα καὶ ἔνθα.

We have already had occasions to mention a few motifs from this passage in connection with the Ciris night episode: Medea’s getting up from her bed (ὀρθωθεῖσα, cf. 209 descendens Scylla cubili), going barefoot (νήλιπος, cf. 212 digitis … primis) and wearing only a light dress (οἰέανος, cf. 252 tenui … crocota). In addition to these not very compelling parallels, there is one that is specific enough: when passing over the threshold (οὐδόν), Medea is stopped by shame (αἰδοῖ ἐεργομένη) and pauses in the forecourt (μίμνεν ἐνὶ προδόμῳ θαλάμοιο), and similarly Scylla, when heading towards the threshold of her father’s chamber (216 qua se ad patrium tendebat semita limen), is seized by fear (214 demptae subita in formidine uires)⁴⁵ and pauses in the forecourt (217 uestibulo in thalami paulum remoratur). The expression πρόδομος θαλάμου(‐οιο) is not Homeric (the closest parallel, and apparently Apollonius’ model, is Il. 9.473 ἐνὶ προδόμῳ, πρόσθεν θαλάμοιο θυράων); apart from the present context, it is only attested once elsewhere in Apollonius (3.839 ἐν προδόμῳ θαλάμοιο) and once in a fragment of Eu-

 I have discussed in the preceding chapter (§) why the Ciris poet replaces shame (αἰδοῖ) with fear (formidine) as the cause of the heroine’s lingering.

110

3 The night episode (lines 206‒348)

phorion (fr. 26.i.10 Lightfoot παρὰ πρόδομον θαλάμοιο). Likewise, uestibulum thalami occurs nowhere else in Latin poetry. The precise equivalence of the (highly uncommon) Greek and Latin expressions, used in identical narrative situations, is as good a proof of a direct relation between the two contexts as one could possibly expect.⁴⁶ Like Medea (3.662 σῖγα μάλα κλαίει), Scylla weeps quietly (211 pressis … singultibus), and as Medea is nevertheless betrayed by the sound of her sobbing (3.664‒665 τὴν δέ τις ἄφνω | μυρομένην … ἐνόησεν), so is Scylla by that of the opening door (221‒222 surgere sensit anus, sonitum nam fecerat illi | marmoreo aeratus stridens in limine cardo, cf. 3.645 θύρας ὤιξε).⁴⁷ When Chalciope learns of Medea’s distress, she rushes to console her (3.670‒672 διὰ δ’ ἔσσυτο θαμβήσασα | ἐκ θαλάμου θάλαμον δὲ διαμπερές, ᾧ ἔνι κούρη | κέκλιτ’ ἀκηχεμένη); and so does Carme (223 corripit extemplo fessam languore puellam). Chalciope and Carme ask Medea and Scylla, respectively, similar questions about the cause of their distress, as we saw in the preceding chapter (2§12). Later on we shall return to these first speeches; at the moment we may focus on the reaction they produce. Medea’s cheeks blush with shame, and because of shame she is unable to speak for some time (3.681‒682 τῆς δ’ ἐρύθηνε παρήια· δὴν δέ μιν αἰδὼς | παρθενίη κατέρυκεν ἀμείψασθαι μεμαυῖαν). Scylla, as we remember, does not experience shame, and here, too, her cheeks are said not to blush, but to be wet with tears (253 genis rorantibus, cf. 3.673 δάκρυσιν ὄσσε πεφυρμένα);⁴⁸ like Medea, she does not answer at once, but the reason is that Carme wants her first to come inside (255‒256 nec tamen ante ullas patitur sibi reddere uoces, | marmoreum tremebunda pedem quam rettulit intra). In contrast to Medea who conceals the real cause of her anguish and lies that she is worried for Chalciope’s children (3.688 περί μοι παίδων σέο θυμὸς ἄηται), Scylla confesses that she has no feelings for those she should love (262 nil amat hic animus, nutrix, quod oportet amari). Chalciope and Carme both react to such revelations with fear (3.695‒ 696 τῆν δ’ αἰνῶς ἄτλητος ἐπέκλυσε θυμὸν ἀνίη | δείματι, 283 clade exterrita tristi). Structurally the conversation of Scylla with Carme is not identical to that of Medea with Chalciope: in the former speeches are fewer and longer, in the latter

 Lyne ()  notes this verbal parallel, as well as the more general one between Ciris ‒ and the Apollonian passage beginning at . (p. ), but does not consider the possibility of the Ciris being directly influenced by the Argonautica.  Perhaps the Ciris poet is ‘correcting’ Apollonius: if the heroine sheds tears silently, why should she be heard?  Or does genae here mean ‘eyes’, as Lyne ()  suggests, thus matching Apollonius’ ὄσσε rather than παρήια?

3.6

111

they are more numerous, but also shorter. The parallels discussed above belong mainly to structurally analogous parts of the two conversations (e. g., the passages introducing the first speech, or the first speech itself, or the passages following it); but there are also shared motifs that have, so to speak, changed their position in that structure. So, in the Argonautica Chalciope implores Medea to swear that she will help (3.697‒704): καὶ δ’ αὐτὴ τάδε πάντα μετήλυθον ὁρμαίνουσα, εἴ τινα συμφράσσαιο καὶ ἀρτύνειας ἀρωγήν. ἀλλ’ ὄμοσον Γαῖάν τε καὶ Οὐρανόν, ὅττί τοι εἴπω σχήσειν ἐν θυμῷ σύν τε δρήστειρα πέλεσθαι. λίσσομ’ ὑπὲρ μακάρων σέο τ’ αὐτῆς ἠδὲ τοκήων, μή σφε κακῇ ὑπὸ κηρὶ διαρραισθέντας ἰδέσθαι λευγαλέως· ἢ σοί γε φίλοις σὺν παισὶ θανοῦσα εἴην ἐξ ᾿Aίδεω στυγερὴ μετόπισθεν Ἐρινύς.

And Medea makes the promise that, ὅσσον σθένος ἐστὶν ἐμεῖο, | μή σ’ ἐπιδευήσεσθαι, ἀνυστά περ ἀντιόωσαν (3.716‒717). In the Ciris, on the contrary, it is Scylla who implores Carme to save her (273‒277): quod per te diuum crebros obtestor amores perque tuum memori sanctum mihi pectus alumnae, ut me, si seruare potes, nec perdere malis; sin autem optatae spes est incisa salutis, ne mihi, quam merui inuideas, nutricula, mortem.

In response to which Carme promises (338‒339): meque deosque tibi comites, mea alumna, futuros | polliceor. ⁴⁹ Some further motifs are derived from the second part of the Apollonian night episode, in which Medea is left alone. We have already seen that the motif of tachycardia (345 crebrosque insani pectoris ictus), which can be traced back to Iliad 10, finds a closer parallel in the Argonautica (3.755 πυκνὰ δέ οἱ κραδίη στηθέων ἔντοσθεν ἔθυιεν, 760 ἐν στήθεσσι κέαρ ἐλελίζετο).⁵⁰ At another point, Medea’s indecision as to whether she should give the φάρμακον to Jason or poison herself with it (3.766‒767 φῆ δέ οἱ ἄλλοτε μὲν θελκτήρια φάρμακα ταύρων | δωσέμεν· ἄλλοτε δ’ οὔ τι, καταφθεῖσθαι δὲ καὶ αὐτή),⁵¹ is mirrored in Scylla’s wavering

 Cf. further Eur. Hip. ‒ μόνον σύ μοι, δέσποινα ποντία Κύπρι, | συνεργὸς εἴης, noted by De Gianni ()  n. .  The parallel is noted by Lyne () .  Here it is not explicitly said that she would use the φάρμακον to kill herself, but see . πασσαμένη ῥαιστήρια φάρμακα θυμοῦ and cf. further .‒. It should be noted, though,

112

3 The night episode (lines 206‒348)

whether to use the ferrum bidens for cutting her father’s purple lock or for killing herself (278‒282): nam nisi te nobis malus, o malus, optima Carme, ante in conspectum casusue deusue tulisset, aut ferro hoc (aperit ferrum quod ueste latebat) purpureum patris dempsissem uertice crinem, aut mihi praesenti peperissem uulnere letum.

The protasis of this unreal conditional sentence (nisi … tulisset) also finds a parallel in an adjacent context where Medea expresses an unreal wish that she were dead πρὶν ᾿Aχαιίδα γαῖαν ἱκέσθαι | Χαλκιόπης υἷας· τοὺς μὲν θεὸς ἤ τις Ἐρινὺς | ἄμμι πολυκλαύτους δεῦρ’ ἤγαγε κεῖθεν ἀνίας (3.775‒777): πρίν ~ ante, θεὸς ἤ τις Ἐρινὺς … δεῦρ’ ἤγαγε ~ in conspectum casusue deusue tulisset. The motif of hiding the ferrum in one’s clothes and then revealing it can be linked to the fact that Medea, when eventually deciding to hand the φάρμακον over to Jason, brings it to the rendezvous concealed in her breast-band (3.867, 1013‒ 1014 ἔξελε μίτρης | φάρμακον). Line 280 has, of course, a well-known parallel in Aeneid 6 where the Sibyl utters nearly the same words as she shows the Golden Bough to Charon (406‒407): at ramum hunc (aperit ramum qui ueste latebat) | agnoscas. It has been argued that since the Bough is modelled partly on the magical plant Medea gives to Jason,⁵² the Virgilian context must be the Ciris’ source; but as we have seen,⁵³ Scylla’s ferrum, too, is related to Medea’s φάρμακον, so that no decisive argument on the chronological relation of the two Latin poems can be drawn from just this parallel. To add one further point, the idea that the same object (ferrum, φάρμακον) can be applied to rather different uses links the Ciris context to yet another model. As Lyne comments, “the ferrum which Scylla carries and which seemed a comparatively harmless pair of scissors in 213 must in 280 be thought of – it certainly sounds like – a knife or the like.”⁵⁴ This ambiguity, created by the metonymical use of ferrum, has its source in Catullus’ translation of Callimachus’ Βερενίκης πλόκαμος. Berenice’s lock offers as an excuse for its separation from the queen’s head – which, of course, parallels that of the purple lock from the head of Nisus – the story of Mt. Athos peninsula being dug through by the Persians (66.47): quid facient crines, cum ferro talia cedant? τί πλόκαμοι ῥέξωμεν, ὅτ’ οὔρεα τοῖα σιδήρῳ | εἴκουσιν; The Catullan allu-

that strictly speaking there are two sorts of φάρμακα, one for charming (θελκτήρια) and the other for harming (ῥαιστήρια); the important point is that both are referred to as φάρμακα.  Cf. recently Nelis () ‒.  And cf. Gall () ‒.  Lyne () .

3.7

113

sion is confirmed by the use of peperissem, which has a notable parallel in 66.45 Medi peperere nouum mare: in both cases the producing (parere) of either a ‘new sea’ (nouum mare) or ‘death’ (letum) is conceived of as being effected by ferrum.

3.7 Despite the fundamental indebtedness of the night episode in the Ciris to that in the Argonautica, the conversation between Scylla and Carme has notable differences from that between Apollonius’ characters. The central difference, as we saw in the preceding chapter, is that Carme, unlike her Apollonian counterpart, is able to diagnose Scylla’s lovesickness. In fact, she openly claims to be, as it were, an expert in love (242‒243): nam te iactari [sc. amore], non est Amathusia nostri | tam rudis ut nullo possim cognoscere signo. This is a significant passage. To begin with, Carme’s statement contains an allusion to a context in Catullus where he similarly claims to be not unfamiliar with love (68.17‒20):⁵⁵ multa satis lusi: non est dea nescia nostri, quae dulcem curis miscet amaritiem: sed totum hoc studium luctu fraterna mihi mors abstulit.

As we already have seen (1§3), this Catullan context is modelled on an epigram by Philodemus (AP 5.112) in which the latter deals with the opposition of (love) poetry as a futile pursuit and philosophy as a serious one, a concept that will be crucial for subsequent Latin poetry. Most relevantly, we could find this opposition, though formulated somewhat differently, in the proem of the Ciris, where philosophy (6 longe aliud studium, cf. AP 5.112.6 λωιτέρης φροντίδος) is, in a comparable manner, contrasted with light poetry (19 ludere, cf. 5.112.5 καὶ παίζειν ὅτε καιρός, ἐπαίξαμεν) and love of renown (1 iactatum laudis amore, cf. 5.112.1 ἠράσθην). Remarkably, the metaphorical use of iactari to describe the effect of love in 241 si alio quouis animi iactaris amore (resumed in 242 quoted above) points precisely to that context in the very beginning of the Ciris. As I was proposing in the preceding chapter, love can be viewed in the Ciris as a central poetological metaphor; this internal reminiscence in Carme’s speech is, I suggest,

 In addition, the rare antonomasia Amathusia has its source in .. Cf. Peirano () ‒: “Thus, by adapting this Catullan expression, the nurse is simultaneously declaring herself not untrained either as a lover or a love poet. The phrase stands as a self-conscious comment on the Ciris’ place within a tradition of writing about love symptoms in poetry.”

114

3 The night episode (lines 206‒348)

one of the pointers that hint at the existence of such a metaphorical, or even allegorical, level of meaning. And there are at least two more internal references of the same kind that link Scylla with the figure of the narrator. First, the enamoured Scylla is said to ascendere … muros (172), which evokes the narrator’s claim to have dared to ascendere collem (8) of wisdom. Second, Scylla looks at the stars (217‒218 alte | suspicit ad celsi nictantia sidera mundi) in the same way as the narrator does (7 altius ad magni suspexit sidera mundi), as part of his philosophical pursuits. If we assume that all three of these points of contact between Scylla and the figure of the narrator are charged with poetological implications, we encounter an evident paradox. On the one hand, Scylla’s actions prompted by her love for Minos (climbing the wall and gazing at the stars) parallel those of the narrator after his conversion to philosophy. On the other, Scylla’s being in love is spoken of in words that cast her, or so it seems, as a counterpart to the narrator when he was still engaged in the pursuit of renown rather than true knowledge. How is this contradiction to be resolved? The answer can be found, at least partly, in Philodemus’ epigram. As I was suggesting earlier, on a certain level Philodemus subverts, rather than insists on, the opposition of poetry and philosophy: although speaking in the epigram of abandoning love and poetry for philosophy, in reality he finds a way of being both a poet and a philosopher. The Ciris poet obviously follows Philodemus, on the one hand, in stating the opposition, but also, on the other, in effectively ignoring it: he claims to have continued working on the Ciris despite his recent interest in philosophy, in precisely the same way as, despite being a professional philosopher, Philodemus does not stop writing epigrams. The next question to ask is what love stands for on the metaphorical level. It is a difficult question, and at present I would only give a tentative, and necessarily approximate, answer. In the most general way the question can perhaps be answered thus: the experience of love is what makes the composition of poetry possible (or, love stands for a certain experience that is necessary for the composition of poetry). As I shall be arguing, in the Ciris this idea finds expression in the fact that both Carme’s and Scylla’s longer speeches are conceived, as it were, as specimens of poetry inspired by love. Lyne has made an important observation that Carme’s first speech (224‒249: 26 lines) “exactly corresponds in number of lines to Scylla’s reply (257‒82); similarly Carme’s speech of 286‒339 (54 lines) is almost matched by Scylla’s of 404‒58 (55).”⁵⁶ It can be added that both longer speeches are defined as questus (285 questu … complorat anili, 401

 Lyne () .

3.8

115

questu uoluebat inani, cf. 400 lamenta, 462 querimonia), a term that may imply elegy. This, coupled with their length and monologic character, distinguishes them – as, almost, formal pieces of poetry – from the two shorter speeches.

3.8 To resume our argument: in contrast to Chalciope, Carme recognises Scylla’s being in love, and Scylla too, in contrast to Medea, admits that she is and names the object of her passion. It may thus seem that, in contrast to the situation in the Argonautica, there exists a complete mutual understanding between Carme and Scylla. I would, however, suggest that one significant point in their conversation has not been fully appreciated. The expertise in love of which Carme boasts is immediately demonstrated by the allusion to the story of Smyrna’s incestuous passion, which she fears may be similar to what Scylla is experiencing; but otherwise, si concessus amor noto te macerat igni (244), Scylla’s love would be absolutely legitimate. Carme seems to have discussed the matter with her charge before (234‒235 quod saepe petenti | iurabas nihil esse mihi), but in fact her language is so inexplicit that, hardly surprisingly, Scylla does not understand her. Being, unlike Carme, unexperienced in erotic love, Scylla confuses it with familial love, judging the former by the standards of the latter⁵⁷ (259‒264): non ego consueto mortalibus uror amore, nec mihi notorum deflectunt lumina uultus, nec genitor cordi est: ultro namque odimus omnis. nil amat hic animus, nutrix, quod oportet amari, in quo falsa tamen lateat pietatis imago, sed media ex acie, mediis ex hostibus – eheu…

What seems unnatural in the case of familial love – that is, to love a stranger and, all the more so, one’s enemy – is exactly the opposite of what is forbidden in the case of erotic love – that is, to love one’s kin. The conflict of the two kinds of love is not a novel theme in the Ciris. In Apollonius Medea is torn between her feelings for Jason and her sense of filial duty. She also, as we have seen, pretends in her conversation with Chalciope to be concerned not for Jason, but for her sister’s children. The same conflict is at

 Ovid’s Smyrna is more knowledgeable of the difference between the two kinds of love, but she, too, is perplexed by the paradox that scelus est odisse parentem, | hic amor est odio maius scelus (Met. .‒).

116

3 The night episode (lines 206‒348)

play in the Lemnian episode of the Argonautica, a context that the Ciris seems to be specifically alluding to. Like Scylla who admits, literally, to hating all her relatives and friends (ultro namque odimus omnis) and to loving a person from among the besiegers (mediis ex hostibus…), the Lemnian men κουριδίας μὲν ἀπηνήναντο γυναῖκας | … ἐχθήραντες, ἔχον δ’ ἐπὶ ληιάδεσσι | τρηχὺν ἔρον (1.611‒613, cf. 804‒806); both Scylla’s ‘strange love’ (non ego consueto mortalibus uror amore) and the τρηχὺς ἔρος of the Lemnian men endanger the existing social order. Hypsipyle’s more thorough account describes in greater detail – in words that are similar to Scylla’s confession – how this illicit passion affected the men’s relations not only with their wives, but with their entire families (1.813‒819, to quote just a part): οὐδὲ πατὴρ ὀλίγον περ ἑῆς ἀλέγιζε θυγατρός, εἰ καὶ ἐν ὀφθαλμοῖσι δαϊζομένην ὁρόῳτο μητρυιῆς ὑπὸ χερσὶν ἀτασθάλου· οὐδ’ ἀπὸ μητρὸς λώβην ὡς τὸ πάροιθεν ἀεικέα παῖδες ἄμυνον, οὐδὲ κασιγνήτοισι κασιγνήτη μέλε θυμῷ. ἀλλ’ οἶαι κοῦραι ληίτιδες ἔν τε δόμοισιν ἔν τε χοροῖς ἀγορῇ τε καὶ εἰλαπίνῃσι μέλοντο…

It has been argued that Apollonius’ treatment of the theme of love in the Argonautica is informed by philosophical discussions of love’s nature and place in life.⁵⁸ Of the two opposed kinds of love, which can conveniently be termed φιλία and ἔρως, Apollonius seems to have a more positive interest in the former, which – following Aristotle – he presents as a central social virtue. In contrast, the Ciris focuses on erotic love, but likewise, I would suggest, from a philosophical perspective, namely interpreting it as an aspiration towards the transcendent. Minos, the object of Scylla’s love, is in a sense a ‘transcendent’ figure simply by virtue of being a stranger and an enemy. From a different angle, that quality of Minos is further emphasised by two subtexts arguably alluded to in the passage where Scylla reveals to Carme the identity of her beloved (268‒ 272): ille, uides, nostris qui moenibus assidet hostis, quem pater ipse deum sceptri donauit honore, cui Parcae tribuere nec ullo uulnere laedi (dicendum est, frustra circumuehor omnia uerbis), ille mea, ille idem oppugnat praecordia Minos.

 See e. g. Kyriakou (), Mori () ‒.

3.8

117

First, I suggest, this passage alludes to the proem of Lucretius’ fifth book in which Epicurus is acclaimed as a deity (5.7‒12): nam si, ut ipsa petit maiestas cognita rerum, dicendum est, deus ille fuit, deus, inclute Memmi, qui princeps uitae rationem inuenit eam quae nunc appellatur sapientia, quique per artem fluctibus e tantis uitam tantisque tenebris in tam tranquillo et tam clara luce locauit.

The phrase dicendum est is admittedly the only exact verbal correspondence between the two contexts (though it should be noted that it does not appear as a parenthesis elsewhere in Latin hexameter poetry); there are, however, certain structural similarities. Both Scylla and Lucretius are reluctant, though to a different degree, to speak out what they have on their minds; but when they do so, the delay is, as it were, compensated by the emphatic use of the figure of reduplication (ille mea, ille, cf. deus ille fuit, deus). Also, both Minos and Epicurus are characterised in relative clauses which sound, in both cases, very much like a hymnic aretalogy. We have seen a number of times how Scylla is cast as a poet figure, and likewise, I suggest, the purpose of this allusion is to draw a parallel between her relation to Minos and Lucretius’ to Epicurus, which basically is that of a poet to his Muse (moreover, elsewhere Lucretius speaks of his attitude to Epicurus in almost erotic terms, cf. 3.3‒5 te sequor … propter amorem, alluding to 1.15‒16 capta lepore | te sequitur). The second important subtext is Apollonius’ portrayal of Aethalides, whom the Argonauts send as an envoy to the Lemnian women (1.640‒649): τείως δ’ αὖτ’ ἐκ νηὸς ἀριστῆες προέηκαν Αἰθαλίδην κήρυκα θοόν, τῷ πέρ τε μέλεσθαι ἀγγελίας καὶ σκῆπτρον ἐπέτραπον Ἑρμείαο σφωιτέροιο τοκῆος, ὅς οἱ μνῆστιν πόρε πάντων ἄφθιτον. οὐδ’ ἔτι νῦν περ ἀποιχομένου ᾿Aχέροντος δίνας ἀπροφάτους ψυχὴν ἐπιδέδρομε λήθη· ἀλλ’ ἥ γ’ ἔμπεδον αἰὲν ἀμειβομένη μεμόρηται, ἄλλοθ’ ὑποχθονίοις ἐναρίθμιος, ἄλλοτ’ ἐς αὐγὰς ἠελίου ζωοῖσι μετ’ ἀνδράσιν. ἀλλὰ τί μύθους Αἰθαλίδεω χρειώ με διηνεκέως ἀγορεύειν;

We just have seen how the treatment of the conflict between familial and erotic love in Scylla’s speech is based on the Lemnian episode, so a further parallel linking the two contexts does not come as a surprise. Of course, Minos’ association with a sceptre is ultimately derived from the Odyssey (11.569 χρύσεον σκῆ-

118

3 The night episode (lines 206‒348)

πτρον ἔχοντα), but the motif of its being a gift from Jupiter (pater ipse deum sceptri donauit honore), who is actually Minos’ father, has a closer parallel in the Apollonian passage (σφωιτέροιο τοκῆος, though cf. Mosch. Europa 161). Both Minos and Aethalides are endowed with exceptional supernatural powers: the former is invulnerable, the latter possesses an imperishable memory. Another shared feature is the breaking of a digression: in the Argonautica the narrator interrupts his account of Aethalides’ reincarnations and returns to the main story, in the Ciris Scylla courageously stops delaying the moment of naming Minos (circumuehor omnia uerbis, cf. μύθους … διηνεκέως ἀγορεύειν).⁵⁹ It has been argued that Aethalides has a certain poetological significance in the Argonautica,⁶⁰ and I shall return to this Apollonian passage later and discuss it in greater detail, for it does have a crucial bearing on understanding the poetics of the Ciris (6§6). At present, however, I confine myself to observing that both intertextual prototypes of Minos – Epicurus and Aethalides – are philosophers: the former quite literally, the latter as an earlier incarnation of Pythagoras’ soul, a fact unambiguously hinted at by Apollonius.⁶¹ Provisionally the poetological implication can be formulated as follows: love – which, according to my reading, stands in the Ciris for an experience fundamentally necessary for composing poetry – is to be understood along philosophical lines, as an aspiration aiming, in

 All these parallels are admittedly not sufficient to prove a direct connection between the two contexts. It is possible to adduce two further arguments from later Latin poetry (assuming the Ciris predates Virgil). First, at Met. .‒ Ovid speaks of Cadmus and his wife transformed into snakes (nunc quoque nec fugiunt hominem nec uulnere laedunt, | quidque prius fuerint placidi meminere dracones) in a way that evokes both the Ciris and the Argonautica: the verse-ending uulnere laed- is only attested here and in the Ciris (furthermore, the exact collocation qui moenibus is paralleled at Met. ., and occurs only once more in Latin poetry, Sil. Pun. .), whereas the motif of preserving one’s memory in a new body (meminere) points to Aethalides (μνῆστιν). Second, the verse-ending assidet hostis has only three contexts in Latin poetry: in the Ciris, in the Aeneid (.‒ non tempore tali | cogere concilium, cum muros adsidet hostis), and in Valerius’ Argonautica (.‒ cuperem haut tali uos tempore tectis | aduenisse meis, quo me grauis adsidet hostis. | frater enim – sceptri sic omnibus una cupido – | excidium parat). The Virgilian context has also in common with the Ciris the motif of invincibility / invulnerability (.‒ inuictisque uiris gerimus, quos nulla fatigant | proelia nec uicti possunt absistere ferro, cf. nec ullo uulnere laedi); Nelis ()  connects, though without discussion, the entire scene of council (.‒) with the analogous scene in the Lemnian episode (.‒). Valerius’ debt to Virgil is obvious, but the presence of a sceptre (as well as the mention of the Parcae at .) links him with the Ciris; the situation of the Valerian Aeetes is similar to that of the Apollonian Hypsipyle: both decide to receive the Argonauts favourably in fear of an imminent attack.  See Nishimura-Jensen ().  See Ardizzoni ().

3.9

119

some way, at metaphysical reality. I suggest that this rather speculative scheme can be interpreted in terms of the thematic principle of composition: the metaphysical reality in which poetry must be grounded can be identified with the abstract rules of ‘themes’ (or type-scenes).

3.9 As pointed out earlier, Carme’s longer speech (286‒339) is given certain attributes of a formal piece of poetry rather than of a mere remark in a dialogue. In addition to what has already been mentioned, I would call attention to the fact that in the first half of the speech (286‒309) Carme is addressing not her actual interlocutor, Scylla, but Minos and Britomartis, who are absent. As I have proposed to argue, this speech can be seen to exemplify, in a way, the connection between love and poetry. It is, I suggest, because of her experience in love that Carme can produce such a sophisticated text (though it seems unclear on exactly what experience Carme’s expertise is grounded, since there is no mention of explicit details of a love affair in which she would have been involved, perhaps apart from the implied circumstance that she mothered Britomartis by Zeus). The way Carme begins her speech is remarkable for at least two reasons (284‒ 289): insontes multo deturpat puluere crinis et grauiter questu Carme complorat anili: ‘o mihi nunc iterum crudelis reddite Minos, o iterum nostrae Minos inimice senectae, semper ut aut olim natae te propter eundem aut amor insanae luctum portauit alumnae.’

First, Carme explicitly names love (or Amor?) as the ultimate cause of her grief, both past and present.⁶² Second, she emphatically draws an analogy between  The text of the passage is badly transmitted, and Goold’s version is not incontestable. For our present concerns, the most relevant question is of whose luctus Carme is speaking. Standardly luctus is here understood as the grief experienced by Britomartis and Scylla (accordingly, natae and insanae are taken to be datives with portauit). But, as has been realised by Luppe () , luctus must refer to Carme’s own grief for her daughter and charge (thus he inserts 〈nobis〉 to go with portauit, whilst taking natae and insanae as objective genitives with luctum). First of all, luctus is “the expression of grief, mourning, lamentation (esp. over the death of someone loved)” (OLD s.v. a), and this applies to Carme’s situation much more closely than to that of either Scylla or Britomartis. I plan to discuss the textual problems in greater detail and to propose my own solution elsewhere.

120

3 The night episode (lines 206‒348)

Scylla’s and Britomartis’ relationships with Minos, despite the fact that the two girls play exactly opposite roles (both suffer and are eventually to die because of love either for or from Minos). Carme’s appreciation of the repetitiveness of life (note iterum [bis!], eundem, 292 bis iam) can be seen, I believe, as a manifestation of the thematic principle of composition, for what she does as she observes this analogy is indeed “extract general patterns of action from the jumbled disorder of events,”⁶³ that is what a good poet does. And exactly the same happens on the intertextual level, as Carme’s speech – or lament, to be precise – is engaged in a complex interaction with a whole range of subtexts, from Homer to Catullus, all belonging to the epic theme of lamentation.⁶⁴ The most immediate source for the passage quoted above is to be found in Catullus 64, in Aegeus’ farewell speech to Theseus (64.215‒224):⁶⁵ ‘gnate mihi longe iucundior unice uita, gnate, ego quem in dubios cogor dimittere casus, reddite in extrema nuper mihi fine senectae, quandoquidem fortuna mea ac tua feruida uirtus eripit inuito mihi te, cui languida nondum lumina sunt gnati cara saturata figura, non ego te gaudens laetanti pectore mittam, nec te ferre sinam fortunae signa secundae, sed primum multas expromam mente querellas, canitiem terra atque infuso puluere foedans.’

The idea of the repetitiveness of action, which I have just called attention to in the Ciris on the narrative level, can now be observed on that of allusion. Perhaps the most distinctive case of that repetitiveness is the repetition of the vocative reddite ‘returned, given back’ from Catullus (reinforced to the point of tautology by iterum).⁶⁶ This is particularly pointed since Minos does not only reappear in Scylla’s story from Britomartis’, but actually he has already been the (indirect) cause of Aegeus’ death in Catullus. Furthermore, it is often assumed that Aegeus’ farewell speech in Catullus, delivered at the moment of Theseus’ departure for the fight with the Minotaur, is based on the (now lost) analogous speech in Cal-

 Janko () .  On laments in Homer, see Tsagalis (). On lament as, in a sense, an archetypal form of epic, cf. Murnaghan ().  Which is an expression of luctus (.).  Lyne ()  interprets iterum crudelis reddite as “again rendered, caused to be, cruel to me,” and this may be the intended meaning on the surface level; but the Catullan subtext invites to read reddite more literally (though less suitably to the context) as ‘given back to me’.

3.10

121

limachus’ Hecale, spoken by Aegeus on the occasion of Theseus’ departure for the fight with the Marathonian bull (which in some accounts is Minotaur’s father).⁶⁷ In fact, there survives a fragment of a speech addressed by Theseus to Aegeus on the same occasion that features noteworthy parallels with both the Catullan context and that in the Ciris (fr. 17.3‒4 Hollis): …ὑπὸ πάντας ἀέθλους. | τῷ ῥα, πάτερ, μεθίει με· σόον δέ κεν αὖθι δέχοιο. Theseus’ request to send him to fight the bull (πάτερ, μεθίει με) is arguably echoed in Aegeus’ remark that he sends his son to Crete against his will (cogor dimittere; mittam). Likewise, the motif of Theseus’ having been ‘given back’ (reddite) to his father is paralleled by Theseus’ promise that he will come back safely for a second time (αὖθι δέχοιο: the first time being when Theseus came from Troezen). If so, I would suggest that iterum … reddite in the Ciris may directly allude to this αὖθι δέχοιο, thus, once more, stressing the theme of repetitiveness not only on the level of narration, but also on that of allusion. This parallel is additionally supported by the fact that the preceding verse-end (ὑπὸ πάντας ἀέθλους) seems to be behind Carme’s words tam duros passa labores (291);⁶⁸ it has been speculated that Theseus is here referring to a promise of aid ‘in every labour’ by Athena,⁶⁹ but it seems more probable that the point of reference is the labours he performed on his way from Troezen to Athens (thus, both Carme and Theseus would be speaking of their past labours which are expected to guarantee future success).

3.10 In addition to the farewell speech that Aegeus (supposedly) addressed to Theseus when sending him against the Marathonian bull, Callimachus’ poem seems to have featured an analogous speech by Hecale. As has been noted by others, Carme’s assertion that she has many times refused to die in order to make a wedding veil for Scylla (316‒317 cum premeret natura, mori me uelle negaui, | ut tibi Corycio glomerarem flammea luto) finds a remarkable parallel in a fragment of a speech by Hecale that may have been precisely that farewell speech (fr. 49.2‒3 Hollis: ἠρνεόμην θανάτοιο πάλαι καλέοντος ἀκοῦσαι | μὴ

 Hollis () ‒.  Note that with ut duros mille labores | rege sub Eurystheo fatis Iunonis iniquae | pertulerit (Aen. .‒) Virgil renders Apollonius’ Ἄργεΐ οἱ μοῖρ’ ἐστὶν ἀτασθάλῳ Εὐρυσθῆι | ἐκπλῆσαι μογέοντα δυώδεκα πάντας ἀέθλους (.‒).  Hollis () .

122

3 The night episode (lines 206‒348)

μετὰ δὴν ἵνα καὶ σοὶ ἐπιρρήξαιμι χιτῶνα).⁷⁰ Although Hecale speaks of quite a different piece of cloth, the point is the same: both old women express their concern that they may have (or have actually had) to bury a much younger person. The Callimachean context is extremely fragmentary, and it is unclear who Hecale’s addressee (καὶ σοί) is at this point. As Hollis observes (though preferring another solution), it is possible to “refer ‘you too’ (fr. 49. 3) to Theseus: Hecale, having lost her own family, is appalled to hear that her young guest plans to fight the Marathonian bull; she feels that he is certain to succumb.”⁷¹ If so, Hecale would have combined lament for her own family with (anticipatory) lament for Theseus, much in the same way as Carme mourns over both her daughter and, in anticipation, her charge. In addition, I would tentatively suggest two further textual points of contact between Hecale’s and Carme’s speeches. First, μέτρα in line 5 may be a reference to the notion of the span of life,⁷² and thus it would be paralleled in 296 diem potui producere uitae. Second, ἐλπίδες in line 7, “if a single and complete word,” is likely to imply “the frustrated hopes of parents,” which “is one of the commonest themes in epitaphs for those who die young”:⁷³ this finds thematic correspondences in 295 mei spes una sepulcri and 311 tui cum spes integra maneret. Furthermore, Hecale’s speech seems to be a model for the farewell speech of Alcimede in Apollonius’ Argonautica (1.278‒291)⁷⁴ – which, in turn, has parallels with Carme’s speech. The verse-end (πάλαι) καλέοντος ἀκοῦσαι appears to be echoed in 1.278 ἐξειπόντος ἄκουσα (with πάλαι used in 1.283): much as Hecale’s words imply a wish that she should have listened to the call of death, so Alcimede wishes that she had died when she first learnt of the expedition, for then she would have been able to be buried by her son. Line 11 ]λαι χέρες is perhaps to be restored as φίλαι χέρες and thus can be linked with 1.281 φίλαις ταρχύσαο χερσίν. In line 12, with τεκ[ at the beginning and ].ισεμονοικ[ in the middle, the former is perhaps to be supplemented as τέκνον, and in the latter ἐμόν can be singled out, so as to match 1.282 τέκνον ἐμόν. (Note that the two last parallels belong to two consecutive verses in both Apollonius and Callimachus.)

 Lehnus () , Lyne () , Hollis () , Gatti () .  Hollis () .  Pace Hollis () .  Hollis () .  For the sake of convenience, I follow Hollis’ ()  assumption that the Argonautica postdates the Hecale (their actual chronology is of little importance for our present purposes). The similarity between Apollonius’ and Callimachus’ parting scenes has been suggested by Ambühl () ‒, but without a detailed discussion of textual parallels.

3.10

123

Line 13 εμαπαν[ – if restored as ἐμὰ πάντα – may be echoed in 1.283 ἄλλα δὲ πάντα. Finally, line 10 ἡμῖν δὲ κακός can be compared with 1.291 ἐμοὶ κακόν. Alcimede’s farewell speech, which can also be described as anticipatory lament, finds parallels in both parts of Carme’s speech, addressed to Britomartis and to Scylla (1.278‒291): αἴθ’ ὄφελον κεῖν’ ἦμαρ, ὅτ’ ἐξειπόντος ἄκουσα δειλὴ ἐγὼ Πελίαο κακὴν βασιλῆος ἐφετμήν, αὐτίκ’ ἀπὸ ψυχὴν μεθέμεν κηδέων τε λαθέσθαι, ὄφρ’ αὐτός με τεῇσι φίλαις ταρχύσαο χερσίν, τέκνον ἐμόν· τὸ γὰρ οἶον ἔην ἔτι λοιπὸν ἐέλδωρ ἐκ σέθεν, ἄλλα δὲ πάντα πάλαι θρεπτήρια πέσσω. νῦν γε μὲν ἡ τὸ πάροιθεν ᾿Aχαιιάδεσσιν ἀγητὴ δμωὶς ὅπως κενεοῖσι λελείψομαι ἐν μεγάροισι, σεῖο πόθῳ μινύθουσα δυσάμμορος, ᾧ ἔπι πολλὴν ἀγλαΐην καὶ κῦδος ἔχον πάρος, ᾧ ἔπι μούνῳ μίτρην πρῶτον ἔλυσα καὶ ὕστατον· ἔξοχα γάρ μοι Εἰλείθυια θεὰ πολέος ἐμέγηρε τόκοιο. ὤ μοι ἐμῆς ἄτης· τὸ μὲν οὐδ’ ὅσον οὐδ’ ἐν ὀνείρῳ ὠισάμην, εἰ Φρίξος ἐμοὶ κακὸν ἔσσετ’ ἀλύξας.

We may start with the Britomartis section. Alcimede’s explicit death-wish (1.278‒ 280) is paralleled in Carme’s complaint that she lived after her daughter’s death (294 … 296): quid ego amens … diem potui producere uitae? (though, as we have seen, the idea of the span of life may be derived from the Hecale). Alcimede’s assertion that Jason has already fulfilled all his filial responsibilities and his only remaining duty is to bury her (οἶον … ἐέλδωρ) is echoed, though with a slightly different emphasis, in Carme’s address to Britomartis as the one who was her only hope of burial (295 mei spes una sepulcri). Turning now to the Scylla section, it seems possible to find the same context reflected once more, though again alio sensu, in 311 tum, mea alumna, tui cum spes integra maneret (cf. τέκνον ἐμόν and ἔην ἔτι λοιπόν). The motif of the child’s uniqueness (ᾧ ἔπι μούνῳ) is paralleled in 314 sola meae uiuendi causa senectae, though likewise with a contrast: Alcimede says that Jason is the only child she gave birth to, Carme calls Scylla her only reason for staying alive in old age. Both women complain of their poor prospects in the immediate future (νῦν γε… 318 quo nunc … me fata reseruant?). Finally, both mention Ilithyia (the Ciris context, line 326, is badly transmitted). Of course, it is not impossible that all these parallels between Carme’s and Alcimede’s speeches are independently derived from the common source in the Hecale; but it seems more natural to assume that the Ciris is here using both Apollonius and Callimachus (not to mention Catullus). In fact, the Ciris can

124

3 The night episode (lines 206‒348)

also be argued to allude to their Homeric subtext. It has been proposed that the curious fr. 122 (Hollis) ἀπούατος ἄγγελος ἔλθοι may come from a context telling of “Aegeus (or even Hecale) dreading a message that Theseus has succumbed to the Marathonian bull,”⁷⁵ possibly one of the farewell speeches. The fragment has a specific Homeric point of reference, as it is a conflation of two lines from a single episode, Il. 22.438 οὐ γάρ οἵ τις ἐτήτυμος ἄγγελος ἐλθών and 454 αἲ γὰρ ἀπ’ οὔατος εἴη ἐμεῦ ἔπος.⁷⁶ This is one of the most dramatic moments in the Iliad, when Andromache hears the wailing aroused by Hector’s death, but does not know yet its cause, when she realises that something dreadful must have happened, but does not want to suspect the worst. Callimachus clearly expected the reader to recall this situation of suspense. The same (broad) context underlies the Apollonian passage: as has been argued, “similarity of theme [and] verbal reminiscence invite the reader to see Andromache’s tragic situation as the literary backdrop for Alcimede’s pitiful lament.”⁷⁷ For the Callimachean fragment a Virgilian parallel has been adduced,⁷⁸ which, I would argue, is more than just an accidental parallel (Aen. 8.578‒583): ‘sin aliquem infandum casum, Fortuna, minaris, nunc, nunc o liceat crudelem abrumpere uitam, dum curae ambiguae, dum spes incerta futuri, dum te, care puer, mea sola et sera uoluptas, complexu teneo, grauior neu nuntius auris uulneret.’

This is part of Evander’s farewell speech to Pallas, which likewise exploits motifs of lament. Virgil’s wording does indeed sound like a translation of the Callimachean phrase (grauior nuntius = ἀπούατος ἄγγελος), which at the same time points to the etymology of the peculiar adjective (‘that needs to be kept away as able to wound the ears’) and, possibly, to its Homeric background (auris uulneret ‘has [not yet] reached my ears’, ἀπ’ οὔατος εἴη ‘may it never reach my ears’). On a larger scale, the Virgilian context seems to be modelled (if we suppose the Ciris predates the Aeneid) on a passage in Carme’s speech (310‒314):⁷⁹

 Hollis () .  On the misinterpretation of ἀπ’ οὔατος as ἀπούατος, see Hollis () ‒.  Clauss () .  Hollis () .  There are also numerous parallels between Evander’s and Carme’s speeches outside the quoted passages.

3.11

125

‘uerum haec tum non sic grauia atque indigna fuere, tum, mea alumna, tui cum spes integra maneret, nam uox ista meas nondum uiolauerat auris. tene etiam fortuna mihi crudelis ademit, tene, o sola meae uiuendi causa senectae?’

In particular, dum … neu … auris uulneret can obviously be linked with nondum uiolauerat auris, both phrases using the same metaphor of wounding the ears. What is especially remarkable is the fact that the wording of the Ciris appears to allude to both original Homeric contexts: nam … nondum capturing the sense of 22.438 οὐ γάρ οἵ τις ἐτήτυμος ἄγγελος ἐλθών (add 437‒438 ἄλοχος δ’ οὔ πώ τι πέπυστο | Ἕκτορος) and uox ista meas … uiolauerat auris rendering 22.454 αἲ γὰρ ἀπ’ οὔατος εἴη ἐμεῦ ἔπος (add perhaps 451 αἰδοίης ἑκυρῆς ὀπὸς ἔκλυον). If my arguments in the last two sections are valid, Carme turns out, as it were, to be exercising in her longer speech the technique of composition by theme, as she models laments for Britomartis and Scylla on (mainly anticipatory) laments in Catullus, Apollonius, Callimachus and Homer.⁸⁰ Of course, we may ask whether the thematic nature of Carme’s speech is actually intended to comment on her character, thus presenting her as a poet figure, or is merely a result of the Ciris author’s own compositional practice. Two arguments can be advanced in support of the former supposition. First, as will be argued in Chapter 5, Scylla’s monologue, which is in a sense a response to Carme’s longer speech, also features poetological connotations, perhaps in a less ambiguous way. Second, as I have already pointed out, Carme manifests her ability to draw analogies – analogical thinking being the cognitive basis of composition by theme – on the narrative level too. It is this second point that I want to develop further.

3.11 The central point of the analogy established by Carme is that amor Minois (either subjective or objective genitive) has ruined both Britomartis and Scylla; and

 In addition to the Homeric allusion discussed above, I would point out that Carme’s speech can be compared with Andromache’s farewell speech to Hector in Iliad  (‒); to list just some points of contact: like Carme, Andromache has lost all her relatives, and Hector is now her only support; like Carme, she attempts to dissuade Hector from what he plans to do; as in the Ciris Minos is the cause of both Britomartis’ and (in future) Scylla’s death, so in Homer it is Achilles who both has killed Andromache’s relatives and is to kill Hector.

126

3 The night episode (lines 206‒348)

Carme laments over both Britomartis and Scylla in a comparable fashion. This is, of course, a deliberate paradox, for it is difficult to imagine more disparate situations than that of a lover (Scylla) and that of a beloved (Britomartis). The analogy might be attributed to the excessive fancy of the old woman, but it turns out that the Britomartis digression in Carme’s speech plays, in fact, a paradigmatic role as it prefigures Scylla’s fate (301‒305): numquam tam obnixe fugiens Minois amores praeceps aerii specula de montis abisses, unde alii fugisse ferunt et numen Aphaeae uirginis assignant, alii, quo notior esses, Dictynnam dixere tuo de nomine Lunam.

This context finds striking parallels, verbal and thematic, in the passage introducing the description of Scylla’s metamorphosis (487‒489): aeriis potius sublimem sustulit alis, esset ut in terris facti de nomine ciris, ciris Amyclaeo formosior ansere Ledae.

Both Britomartis and Scylla are miraculously rescued from imminent death in the sea (whose cause would have been Minos) and, by being either deified (the former) or metamorphosed (the latter), become immortal. Both accounts manifest aetiological and etymological concerns (note de nomine): Britomartis’ leap into the sea explains the cult of Aphaea (as well as Diana’s alias Dictynna), Scylla’s end accounts for the (mythological) existence of the ciris. The analogy explicitly stated by Carme is not only at play, as we have just seen, on the intratextual plane, but it also has an intertextual dimension. The Britomartis excursus in the Ciris is known to be largely indebted to Callimachus’ account of Britomartis in the hymn to Artemis (3.189‒191, 195‒199):⁸¹ ἔξοχα δ’ ἀλλάων Γορτυνίδα φίλαο νύμφην, ἐλλοφόνον Βριτόμαρτιν ἐύσκοπον· ἧς ποτε Μίνως πτοιηθεὶς ὑπ’ ἔρωτι κατέδραμεν οὔρεα Κρήτης. … μέσφ’ ὅτε μαρπτομένη καὶ δὴ σχεδὸν ἥλατο πόντον πρηόνος ἐξ ὑπάτοιο καὶ ἔνθορεν εἰς ἁλιήων δίκτυα, τά σφ’ ἐσάωσαν· ὅθεν μετέπειτα Κύδωνες νύμφην μὲν Δίκτυναν, ὄρος δ’ ὅθεν ἥλατο νύμφη Δικταῖον καλέουσιν…

 Dal Zotto () ‒, Skutsch () ‒, Lyne () ‒.

3.11

127

In her turn, as has been observed, Callimachus’ Britomartis, “who is celebrated here for preserving virginity by leaping into the sea to escape the pursuing Minos”, finds a remarkable counterpart in the figure of “the nymph Asteria, who in Callimachus’ fourth hymn is transformed into the island of Delos, an event which […] is ultimately dependent upon the nymph’s leap into the sea to flee the pursuing Zeus.”⁸² For an account of Asteria’s leap into the sea Depew cites 4.244‒248, but a more relevant version is given earlier in the hymn (4.36‒38): …οὔνομα δ’ ἦν τοι ᾿Aστερίη τὸ παλαιόν, ἐπεὶ βαθὺν ἥλαο τάφρον οὐρανόθεν φεύγουσα Διὸς γάμον ἀστέρι ἴση.

The verse-end 4.37 ἥλαο τάφρον is evidently meant to recall 3.195 ἥλατο πόντον⁸³ (and vice versa). In addition, it should be pointed out that both contexts are explicitly concerned with etymologies: the former explains why Britomartis is called Dictynna (as well as why the mount is called Dicte), the latter why Delos used to be called Asteria. I would suggest that the analogy between Britomartis and Scylla in the Ciris may be intended specifically to imitate that between Britomartis and Asteria in Callimachus’ hymns. On the one hand, the Ciris poet seems to acknowledge the link between the two Greek contexts by drawing a parallel between his own Britomartis, fugiens Minois amores, and Callimachus’ Asteria, φεύγουσα Διὸς γάμον; by contrast, the hymn to Artemis speaks of Minos’ chasing (3.191 κατέδραμεν, 194 διωκτύν) rather than of Britomartis’ fleeing. On the other hand, as I shall be arguing in a later chapter (5§4), Scylla has further significant points of contact with the Delos of Callimachus’ hymn. Finally, Ovid’s curious version in which Scylla is said to leap into the sea on her own accord so as to pursue Minos’ ship possibly attests to his noticing both the parallelism between Britomartis and Scylla and its model in Callimachus (Met. 8.142 insilit undis is an exact metrical equivalent of ἥλατο πόντον, used likewise at a verse-end).

 Depew () .  As Bornmann ()  observes, Callimachus’ use of a direct accusative with ἅλλεσθαι is striking, for normally that construction would mean ‘to leap over’ rather than ‘to leap into’; this shared irregularity reinforces the link between the two contexts. Furthermore, . Διὸς δ’ ἀνθείλετο πόντον seems likewise to evoke . ἥλατο πόντον; on this kind of sound allusion in Hellenistic poetry, cf. Spanoudakis () ‒.

4 Heroic deeds (lines 349‒385) 4.1 In contrast to the night episode, the next section of the Ciris begins, in a typically epic manner, at dawn. In fact, while the preceding narrative plays on the conventions of the ‘theme of ambush’, which from a Homeric perspective may look rather marginal, the new episode can be said, in a sense, to explore the most archetypal of epic themes, that of the hero killing the monster.¹ Nevertheless, much as the ultimate model of the Ciris’ central episode in the seemingly unconventional Doloneia casts the Latin poem as, almost, an anti-epic, so does the subversive (and implicit) treatment of the most traditional epic theme here, in the episode which is by far the shortest self-contained narrative unit in the whole poem. Moreover, the subversive character of the episode is also evidenced in the fact that, although awaking expectations of a conventional epic narrative by the way it begins, it has – contrary to the standard epic practice – no time reference at the end, so that the reader is left without a clue about the duration of the events described.²

4.2 Perhaps the most obvious subtext in the present section of the Ciris is the wedding song performed by the Parcae in Catullus 64 (322‒381). To begin with, it likewise opens with a mention of the evening star (228‒229): adueniet tibi iam portans optata maritis | Hesperus. When dealing with the motif of the ambivalent attitude to Vesper in the preceding chapter, I did not discuss why in the Ciris, in contrast to the men of the Callimachean fragment, girls shun the star when it appears in the evening and welcome it in the morning. As Lyne plausibly suggests, the Ciris seems to allude to “an epithalamial situation (contrast 350 where the originally epithalamial motif – Oeta – was inert): these girls are comparable to Catullus’ chorus of innuptae in poem LXII who state and explain their dislike of the evening star (lines 21 f.).”³ This is, however, a pointedly ironic allusion, since Scylla actually would be only too glad to see the evening star at her

 On its Indo-European roots, see Watkins ().  Cf. Bartels () .  Lyne () .

4.2

129

wedding to Minos. By way of a sort of ring composition, the Parcae’s song ends with a reference to the next morning (64.376‒380):⁴ non illam nutrix orienti luce reuisens hesterno collum poterit circumdare filo, anxia nec mater discordis maesta puellae secubitu caros mittet sperare nepotes.

It is perhaps no chance coincidence that Scylla is said to follow her nurse’s instructions early in the morning (349 postera lux ubi, 353 praeceptis paret uirgo nutricis), that is at the same time as the nurse is expected to visit the bride after her first wedding night in Catullus. More certainly, the bride’s mother’s (relieved) concern about (not) having grandchildren is echoed in Scylla’s complaint that her father does not care for sharing grandchildren with Jupiter (360‒361 flet maesta parentem, | cum Ioue communis qui nolit habere nepotes). From singing of the bliss of married life the Parcae proceed to foretelling the future exploits of the offspring which is to result from that union: this turns out to be, as it were, a concise digest of the Iliad and the Trojan cycle. Achilles’ glory will culminate in the sacrifice of Polyxena at his tomb (64.368‒370): alta Polyxenia madefient caede sepulcra; quae, uelut ancipiti succumbens uictima ferro, proiciet truncum summisso poplite corpus.

We may begin by pointing out a remarkable detail, whose significance will become apparent later on: the sacrifice of Polyxena is spoken of in such a way as to recall the description of Achilles’ war killings only a few lines earlier (64.355 prosternet corpora ferro and 360 alta tepefaciet permixta flumina caede).⁵ Turning back now to the Ciris context, we may note that although the motif of sacrifice is employed there in a completely different way (to begin with, it is not a human sacrifice), the wording is certainly indebted to Catullus, the collocation uictima ferr- being unattested elsewhere in Latin poetry (365‒ 368): quin etiam castos ausa est corrumpere uates, ut, cum caesa pio cecidisset uictima ferro, essent qui generum Minoa auctoribus extis iungere et ancipitis suaderent tollere pugnas.

 Cf. Ruiz Sánchez () ‒.  Note the pointed use of altus in two opposite senses: ‘high tomb’ vs. ‘deep stream’.

130

4 Heroic deeds (lines 349‒385)

In Catullus, of course, the sacrifice of Polyxena inaugurates the end of the Trojan war, but it can also be viewed as her union with Achilles (note that this incident forms part of an epithalamium; straight after it the Parcae resume: 64.372 quare agite optatos animi coniungite amores). The Ciris exhibits a similarly intricate conjunction of the motifs of sacrifice, ending war, and marriage. In turn, Catullus’ account of the sacrifice of Polyxena has been demonstrated to depend on the famous treatment of the sacrifice of Iphigenia in Lucretius;⁶ it is easy to see that Catullus’ Lucretian subtext has also found its way into the Ciris. The Lucretian passage, too, features a connection, and a more explicit one, between sacrifice and marriage (1.98‒101): sed casta inceste nubendi tempore in ipso hostia concideret mactatu maesta parentis, exitus ut classi felix faustusque daretur. tantum religio potuit suadere malorum.

To begin with textual affinities, the form nubendi occurs in poetry, apart from the present context and that in the Ciris (354), only once in Propertius (4.4.88, in the Tarpeia elegy which is arguably indebted to the Ciris ⁷) and once in Ovid (Her. 20.110 ipso nubendi tempore, obviously alluding to Lucretius). Furthermore, the collocation maesta parent- at verse-end is only attested, in addition to the Lucretian context and Ciris 360 flet maesta parentem, once more in the Ciris (235) and once in Cicero’s Aratea (fr. 31.2), in a passage reminiscent of Iphigenia’s situation, with maesta used of Andromeda and parentis of Cassiopeia who, like Agamemnon, sacrificed her own daughter.⁸ These lexical points of contact are supplemented by larger-scale thematic parallels. First of all, both Lucretius and the Ciris are concerned with a father and daughter relationship, and in both cases the daughter suffers because of her father’s interests. Also, both contexts present marriage as in some way a possible even if unrealised alternative to the daughter’s eventual death. Scylla is thus clearly paralleled with Lucretius’ Iphigenia and Catullus’ Polyxena, girls who, quite literally, have fallen victim – the central point for Lucretius  See Skinner (), Morisi ().  Hutchinson ()  considers that possibility, and cf. above §.  Note that in both Lucretius and Cicero the phrase is immediately preceded by a verbal noun of the fourth declension: mactatu / aspectum (maesta parentis). Cf. further Arat. fr.  hanc autem illustri uersatur corpore propter | Andromeda, aufugiens aspectum maesta parentis with Lucr. .‒ et maestum simul ante aras adstare parentem | sensit et hunc propter ferrum celare ministros | aspectuque suo lacrimas effundere ciuis. For further cases of allusion to Cicero’s Aratea in Lucretius, cf. Edmunds () and esp. Gee () ‒, ‒.

4.3

131

– to religious superstition. We may remember that Nisus’ unbending confidence, too, has its ground in a prophecy. Yet at the same time, it is none other than Scylla herself who exploits religion for her own ends. Lucretius’ condemnatory dictum (tantum religio potuit suadere malorum), followed by a warning to Memmius against the uates who intimidate people (1.102‒103 tutemet a nobis iam quouis tempore uatum | terriloquis uictus dictis desciscere quaeres), can be fittingly applied to Scylla: she frightens her fellow citizens with superstitious fears (363 diuum terret formidine ciues ⁹) and bribes the uates, so that they would persuade (suaderent) Nisus, by means of false prophecies, to stop the war and marry her to Minos. Indeed, not only can Nisus be blamed for what is to happen to his daughter, but Scylla, too, consciously (even if unwillingly, in a sense) acts in such a way as to bring about her father’s death.

4.3 We may now consider another context exploring the theme of sacrifice, although it comes from near the end of the poem. After Scylla has been transformed by Amphitrite into the ciris, Jupiter decides to give life back to Nisus as well (520‒527): nec tamen hoc ipsum poena sine: namque deum rex, omnia qui imperio terrarum milia uersat, commotus talem ad superos uolitare puellam, cum pater extinctus caeca sub nocte lateret, illi pro pietate sua (nam saepe nitentum¹⁰ sanguine taurorum supplex resperserat aras, saepe deum largo decorarat munere sedes) reddidit optatam mutato corpore uitam.

To begin with, the Ciris clearly alludes to Catullus 76 (26 o di, reddite mi hoc pro pietate mea: the collocation pro pietate is not attested elsewhere in Latin poetry), a poem in which, however, pietas is conceived of in a completely different way, in terms of morality rather than ritual.¹¹ A similarly subversive effect is produced

 Contrast Lucr. . aspectuque suo lacrimas effundere ciuis. Furthermore, as Lyne ()  points out, diuum … formidine is a Lucretian phrase (., .).  The transmitted reading saepe uidemus has been variously emended; in Kayachev (forthcoming-b) I propose uidendus, in view of the Lucretian parallel (. saepe uideri).  Cf. e. g. Henry () : “It is clear that to Catullus […] pietas is a quality that is shown in social life, in a man’s relations with his fellows”; and see, recently, Perotti () ‒.

132

4 Heroic deeds (lines 349‒385)

by a Lucretian subtext that directly attacks the understanding of pietas implied in the Ciris (5.1198‒1202):¹² nec pietas ullast uelatum saepe uideri uertier ad lapidem atque omnis accedere ad aras nec procumbere humi prostratum et pandere palmas ante deum delubra nec aras sanguine multo spargere quadrupedum nec uotis nectere uota.

Such a traditional notion of pietas, Lucretius proceeds to explain, is grounded in the fear ne quae forte deum nobis inmensa potestas | sit, uario motu quae candida sidera uerset (5.1209‒1210). In turn, the Lucretian passage has been suggested to refer to a fragment of Empedocles speaking of a golden age that did not know animal sacrifices (fr. 128.8‒10):¹³ ταύρων δ’ †ἀκρίτοισι† φόνοις οὐ δεύετο βωμός, ἀλλὰ μύσος τοῦτ’ ἔσκεν ἐν ἀνθρώποισι μέγιστον, θυμὸν ἀπορραίσαντας ἐέδμεναι ἠέα γυῖα.

What we have in the Ciris thus turns out to be a ‘window’ allusion through Lucretius to Empedocles: the Ciris replaces Lucretius’ quadrupedum with taurorum, to match ταύρων, and instead of an abstract inmensa potestas, it speaks of the king of the gods (deum rex), which, again, is precisely the language used by Empedocles earlier in the fragment (fr. 128.2‒3 οὐδὲ Ζεὺς βασιλεύς … ἀλλὰ Κύπρις βασίλεια). Finally, Nisus’ piety (526 saepe deum largo decorarat munere sedes) is pointedly contrasted with the bloodless sacrifices with which the narrator claims to have honoured the Muses (94‒95 altaria postis | munere saepe meo inficiunt). Another relevant Empedoclean fragment provides an explanation of why animal sacrifices are impious (fr. 137): μορφὴν δ’ ἀλλάξαντα πατὴρ φίλον υἱὸν ἀείρας σφάζει ἐπευχόμενος μέγα νήπιος †οἱ δὲ πορεῦνται† λισσόμενον θύοντες· †ὁ δ’ ἀνήκουστος† ὁμοκλέων σφάξας ἐν μεγάροισι κακὴν ἀλεγύνατο δαῖτα. ὡς δ’ αὔτως πατέρ’ υἱὸς ἑλὼν καὶ μητέρα παῖδες θυμὸν ἀπορραίσαντε φίλας κατὰ σάρκας ἔδουσιν.

 The Lucretian parallel is noted in Salvatore () , but without discussion.  Gale () .

4.4

133

The sacrificial animal will necessarily be a reincarnation of a human soul; what is more, it may even happen to be that of a close relative. It is hardly a coincidence that the Ciris context is concerned with metamorphosis, which is the mythological equivalent of metempsychosis; in particular the phrase mutato corpore can be linked with μορφὴν δ’ ἀλλάξαντα¹⁴ (even transformed into the sea eagle, Nisus continues to do evil to his own daughter). It has also been pointed out that the fragment “recalls the great family murders of tragedy, and in particular is in the opening lines strongly reminiscent of Agamemnon’s sacrifice of his daughter as told by the chorus in Aeschylus”;¹⁵ and it has been plausibly suggested that this fragment may have inspired Lucretius’ use of the Iphigenia story.¹⁶ Looking back at the account of Scylla’s attempts to persuade her father to stop the war and marry her to Minos, we may point out that the motif of (ineffective) supplication – which is barely touched upon in Lucretius (1.92 muta metu terram genibus summissa petebat: Iphigenia is silent) – is developed in the Ciris quite explicitly (355 temptantur patriae submissis uocibus aures), thus paralleling both Empedocles (fr. 137.3 λισσόμενον, ἀνήκουστος) and Aeschylus (Ag. 228 λιτὰς δὲ καὶ κληδόνας πατρῴους).¹⁷

4.4 In an earlier chapter I suggested that the Ciris can be seen to explore the tension between an Empedoclean and a Lucretian tradition in ancient epic (2§8). At the core of this tension, as I pointed out, is a fundamental disagreement on the nature of the soul: whether it is immortal or mortal. Later on we shall analyse how this dilemma can be projected onto the figure of the poet (that is, what it means for the poet to have an immortal as opposed to a mortal soul: see below 6§6), whereas at present I shall offer some thoughts on the implications of this fundamental issue for the epic universe that the poet creates. As a starting point I choose what may appear too remote a perspective, but I hope that such a degree of abstraction will eventually prove useful and not far-fetched. As I suggested in

 The precise collocation mutato corpore has, somewhat unexpectedly, only three occurrences in Latin poetry: apart from the Ciris, it is Lucr. . and Verg. Georg. .. The Lucretian context explicitly attacks the doctrine of metempsychosis, but also the Virgilian one can be argued to explore Empedoclean themes (cf. Pontani [] on cosmological interpretations of the Homeric Proteus).  Wright () ; cf. Campbell () .  Furley () .  Cf. further Gale () on Aeschylus’ Agamemnon as an intertext for Catullus .

134

4 Heroic deeds (lines 349‒385)

passing at the beginning of this chapter, the present section of the Ciris can be read through the lens of the most archetypal epic theme, ‘the hero kills the monster’; let us now focus on this theme and sketch its evolution, albeit in very general outline. This is clearly not the place to rehearse Watkins’s detailed exploration of the underlying formula in which he demonstrates that not only the theme itself, but more specifically its ‘material’ linguistic form can be traced back to a prehistoric Indo-European epic tradition. Yet Watkins also offers a generalising interpretation of the theme, which it will be convenient to expand upon: Why does the hero slay the serpent? What is the function of this widespread if not universal myth, or put another way, what is its meaning? […] The dragon symbolizes Chaos, in the largest sense, and killing the dragon represents the ultimate victory of Cosmic Truth and Order over Chaos. As a part of the Frazerian ‘dying god’ myth, it is a symbolic victory of growth over stagnation or dormancy in the cycle of the year, and ultimately a victory of rebirth over death.¹⁸

As Watkins shows, the basic formula of the theme (‘the hero kills the monster’) has numerous variants (such as, most generally, ‘the monster kills the hero’, ‘the hero kills the (anti‐)hero’, or ‘the hero overcomes death’), many of which may synchronically be represented within a given tradition. I would, however, suggest that, although most such variants can be postulated at an Indo-European level, in different periods and (or) genres some of them may be conspicuously more prominent (and ideologically charged) than others. In terms of this archetypal theme, the development of the classical epic – or, more generally, literary – tradition can be described, though hardly with microscopic precision, as a gradual internalisation of the conflict between order and chaos (such a chronological progress can be attributed both to the actual evolution of mythology and literature and to the internal world-history represented in mythology and literature). In primeval times the conflict is manifested on a cosmic level, as gods and heroes fight against monsters (for example, Gigantomachy, the labours of Heracles). At a later point, when all monsters have been defeated, it shifts to a civilisational level and can be seen in the opposition of Greeks and barbarians (the Trojan war, the Persian wars). Then it penetrates the confines of polis and even family (tragedy). Finally the conflict explodes within the soul of a single person, as a fight of virtues against vices (the fullest and most explicit treatment of this internal con-

 Watkins () .

4.4

135

flict can be found, of course, in Prudentius’ Psychomachia, but it is already adumbrated in Virgil’s Aeneid if not in Apollonius’ Argonautica ¹⁹). The dynamics of this process are by no means strictly linear, especially when looked at from a closer distance. But on a larger scale it certainly does have a linear constituent, which is not difficult to account for: as man conquers manifestations of chaos in what is more distant from himself (first of all, the forces of nature), but realises that chaos still persists (ultimately, in the form of death), he gradually moves the firing line closer and closer to his own self. On a more detailed scale, however, other driving forces can be seen as well. Most importantly, the objective tendency to internalisation of the conflict is at any given moment opposed by the subjective will to its externalisation (it is always easier to fight against what is alien than what is close to oneself). At a certain moment, especially when the straightforward formula ‘the hero kills the monster’ is replaced by the ambiguous ‘the hero kills the anti-hero’ (it can always be questioned who is the actual hero and who the anti-hero), there comes a perception that every single victory over a manifestation of chaos is in danger of creating at the same moment a new manifestation of chaos (which finds expression in the widespread mythologeme ‘the dragon-killer becomes a dragon himself’). This is precisely the paradox that lies at the heart of many Greek tragedies, and perhaps most prominently of Aeschylus’ trilogy, in which every killing has to be avenged by a new killing. It may be a misleading generalisation from the viewpoint of tragedy itself, but from a wider perspective of the ancient literary tradition it seems possible to say that at least one of the ways in which such a chain of killings can be interrupted in tragedy is by substituting an animal sacrifice (a Girardian scapegoat) for another killing, that is, in a sense, by externalising the archetypal conflict between order and chaos when it has invaded human society. It is, I suggest, at this point in the evolutional trajectory of the basic epic theme that Empedocles, with his insistence that animal sacrifice is no better than murder, can be situated. After Empedocles, at least potentially, every killing, whether of a human or an animal, does not so much restore order as produce chaos.²⁰

 Cf. P. Hardie ()  n. : “It is a curious reflection that a very early (Theagenes?) allegorization of the theomachy in the Iliad as a battle between psychological forces (schol. B ad Il. .) already supplies a fully developed example of psychomachy, whose treatment by Prudentius is taken by C. S. Lewis, The allegory of love (Oxford, ), to mark the beginning of a fully medieval approach to allegory.”  Another consequence of this general tendency is that traditional epic killings in war come now to be assimilated to human sacrifices, as we have noted earlier for Catullus  and as Panussi () demonstrates for Virgil’s Aeneid.

136

4 Heroic deeds (lines 349‒385)

In contrast to the traditional epic hero who, for his heroic deeds (mainly killings) and death, is rewarded with an afterlife in the fame perpetuated by poetry, Empedocles’ hero, Pythagoras, has overcome death by conquering ignorance with knowledge (fr. 129; cf. fr. 112). The central point of this knowledge is, of course, the doctrine of metempsychosis. Empedocles can thus be said to have paved the ground for a psychomachy; and although Apollonius’ Argonautica, which in a way develops the Empedoclean tradition, is anything but a psychomachy, it is certainly concerned with the issues of the immortality of the soul²¹ and of ethics as the proper playground for resolving conflicts.²² Nevertheless, as I insisted earlier, despite the radical difference in means by which the Homeric hero as opposed to the Empedoclean achieves immortality, the two epic universes can be seen as fundamentally compatible, for in both the ultimate end is to overcome death and oblivion. In traditional Greek epic the conflict between order and chaos was conducted on two levels: on the cosmic level, with Zeus securing eternal dominance, and on the civilisational level, with heroes fighting against monsters and enemies and thus achieving undying fame and sometimes deification. Now it has expanded to a third plane, that of intellectual or spiritual warfare; but the goal, immortality, has remained the same. This fundamental analogy uniting the three levels is what makes it easy to take one of them as a figure for the two others (hence the physical and moral allegories of the Homeric poems). It is in this situation that Lucretius comes into play. In a paradoxical way, Lucretius’ hero, Epicurus, overcomes (the fear of) death by surrendering to it (note 3.869 mortalem uitam mors cum immortalis ademit; and of course, in an Epicurean universe chaos rules over order). In itself, this might not have been too radical a challenge to the epic tradition, for, in a sense in a comparable way, the Homeric hero acquired poetic immortality precisely by dying a heroic death. It can thus be speculated that had Lucretius entered the epic tradition before Empedocles, he might have fitted within it more organically; but as it was, with the doctrine of metempsychosis he had to reject the entire epic worldview retrospectively conceptualised (arguably, by Apollonius in the first place) so as to conform with Empedocles’ theories. Be that as it may, what best illustrates the subversive potential of Lucretius’ poetry in respect of the traditional epic paradigm is the positive application of the example of the Giants to Epicurus’ attack on

 Most conspicuously, Aethalides is almost explicitly presented as a previous incarnation of Pythagoras (.‒), cf. below §.  See, e. g., Mori () on Apollonius’ engagement with Aristotle’s moral philosophy; cf. also Kyriakou ().

4.5

137

religion.²³ In the war that is to be fought in one’s soul, or rather in one’s mind, Lucretius is to Empedocles not an ally, but an adversary.

4.5 It is, of course, difficult to say how much of all this could be intuited from inside the ancient epic tradition, and we can be certain that whatever was intuited would have been phrased in different terms. I would, however, argue that an understanding along such lines of what is the essence of epic action underpins the narrative in the present section of the Ciris; the way Scylla (together with Carme) acts here is intended to evoke, by means of implicit and sophisticated allusions, several layers in which the basic epic conflict can take place. On the literal level, Scylla’s action – as she attempts, by hook or by crook, to make Nisus do what she wants – can be described as unscrupulous ideological warfare. The arsenal employed by Scylla is indeed exceptional: from straightforward pleas addressed to Nisus she turns to imploring his friends and frightening the Megarians, she invents omens and even bribes augurs, while at the same time Carme performs the magical ritual of defixio to subdue Nisus’ will. It is no chance coincidence that Scylla’s failure to persuade Nisus (378 uerum ubi nulla mouet stabilem fallacia Nisum … 382 purpureumque parat rursus tondere capillum) is described in such a way as to echo her earlier failure to conquer her own passion (181 atque ubi nulla malis reperit solacia tantis … 185‒186 crinem de uertice sacrum | furtiue arguto detonsum mitteret hosti).²⁴ Scylla’s ‘attack’ on Nisus is thus paralleled with the psychological conflict of love and shame – a psychomachy in a nutshell²⁵ – in the earlier passage. As we have seen, an important subtext for the present episode is found in Epicurus’ intellectual campaign against religion, which Lucretius pointedly depicts as a traditional epic fight against a monster, though also as a Gigantomachy (1.62‒101). Yet it is a subtext that is impossible to interpret in a straightforward

 See Gale () ‒. Though Lucretius’ imagery is ambiguous, for in fact he also presents religion as a (Scylla-like?) monster (.‒): quae caput a caeli regionibus ostendebat | horribili super aspectu mortalibus instans.  Verg. Georg. . uerum ubi nulla fugam reperit fallacia is thus arguably a conflation of  atque ubi nulla malis reperit solacia tantis and  uerum ubi nulla mouet stabilem fallacia Nisum; though contrast Munari () ‒.  Of course, this potential is not realised there, as we saw earlier (§), but it is certainly present; though Scylla yields to her passion without giving it a proper fight (or at least that fight is not narrated), the reader is aware that such a conflict plays a prominent part in the Ciris’ models.

138

4 Heroic deeds (lines 349‒385)

manner: on the one hand, by a strikingly precise Lucretian allusion earlier in the poem (143‒144 lasciuit et extra | procedit longe, cf. Lucr. 1.73‒74 peruicit et extra | processit longe), Scylla is, as it were, assimilated to Epicurus the conqueror of religion, whereas here, on the other hand, she sides with religion (though, again, also falls victim to it). I would tentatively suggest, even if this may be an oversimplification, that Scylla’s ambiguous relation to the figure of Epicurus self-reflectively represents the situation of a post-Lucretian poet who can no longer take the mythological world-view at face value, but still has to follow its conventions in poetry.

4.6 On the level of subtexts, as we already have seen, the narrative of the present episode is projected against the theme of (human) sacrifice as the most controversial type of epic killing, which can be considered both an act of retribution or expiation and a new crime at the same time. The sacrifices of Iphigenia and Polyxena, clearly paralleling each other, exemplify this duality in a complementary manner: the former is Agamemnon’s crime, the latter is a revenge for Achilles’ death. From a certain point of view, it is in response to this inherent ambiguity that in the Ciris Scylla is cast as both the victim and the perpetrator of a sacrifice. It may not be irrelevant to remember that the whole story began at a religious festival when Scylla profaned Juno’s shrine during the performance of a sacrifice. As I have argued, Scylla’s at first sight almost innocent offence, for which she nevertheless is severely punished by Amor, is linked by means of allusion to several cases of an exceptional crime (though, again, sometimes she could be said to parallel the victim rather than the offender). In particular, Scylla’s ball can be connected to the ball Aphrodite bribes Eros with in the Argonautica; Eros’ ball seems in turn to be modelled on the magic mirror with which the Titans entice Dionysus in a fragmentary Orphic poem.²⁶ As I also pointed out, the same Orphic subtext appears to be at play in the Absyrtus episode of the Argonautica, and now I would tentatively suggest that Scylla’s actions may be intended to evoke the murders of both Absyrtus and (less certainly) Dionysus. To start with, let us consider parallels between Apollonius and the Orphic narrative. Like Dionysus, Absyrtus is lured out of safety by a trick (Clem.

 It is not certain that all the details of the Dionysus myth are derived from the same Orphic source, but it is probable that Apollonius would have been familiar with whatever sources there were.

4.6

139

Protr. 2.17 δόλῳ δὲ ὑποδύντων Τιτάνων, ἀπατήσαντες παιδαριώδεσιν ἀθύρμασιν, cf. Arg. 4.421‒422 ὣς τώ γε ξυμβάντε μέγαν δόλον ἠρτύναντο²⁷ | ᾿Aψύρτῳ, καὶ πολλὰ πόρον ξεινήια δῶρα). Among the presents sent to Absyrtus to entice him is the garment upon which Dionysus lay with Ariadne during their wedding night: not only is this an explicit reference to Dionysus, but also the enticement radiated by the garment (428‒429 οὔ μιν ἀφάσσων | οὔτε κεν εἰσορόων γλυκὺν ἵμερον ἐμπλήσειας) closely resembles that of the mirror looking into which Dionysus proceeded to the place of slaughter. Dionysus was yet a child (εἰσέτι παῖδα ὄντα … ἔτι νηπίαχον ὄντα); Absyrtus is compared with a naïve child (4.460 ἀταλὸς πάις οἷα). Dionysus had transformed into a bull, in some accounts, when he was killed; in the moment of murder Absyrtus is compared to a bull being sacrificed (4.468). The Titans dismembered Dionysus, and in some versions of the Argonautic story Absyrtus, too, is dismembered. Finally, the Dionysian myth explains the origin of mankind, and similarly Absyrtus becomes the eponym for ἀνδράσιν ᾿Aψυρτεῦσιν (4.481). Textual parallels with the Ciris are suggestive at best, but given our poet’s interest in the theme of human sacrifice, such an allusion – to the murder of Absyrtus in the first place, but possibly also to that of Dionysus – would be natural. To begin with, it can be pointed out that the motif of deceit (362‒363 conficta dolo mendacia turpi | inuenit) cannot be derived from Lucretius or Catullus, but, as we just have seen, is paralleled in both Apollonius and the Orphic fragment. The motif of sacrifice is also present in the Ciris as well as in Apollonius, though in different ways (366 cum caesa pio cecidisset uictima ferro, 4.468‒469 τὸν δ’ ὅ γε, βουτύπος ὥς τε μέγαν κερεαλκέα ταῦρον, | πλῆξεν). But perhaps the most telling parallel is established by the motif of spitting three times (371‒373): terque nouena ligans triplici diuersa colore fila ‘ter in gremium mecum’ inquit ‘despue, uirgo, despue ter, uirgo: numero deus impare gaudet’.

Innocent as it may appear here, in the Argonautica Jason is said to lick and spit three times the blood of Absyrtus (4.478‒479 τρὶς δ’ ἀπέλειξε φόνου, τρὶς δ’ ἐξ ἄγος ἔπτυσ’ ὀδόντων, | ἣ θέμις αὐθέντῃσι δολοκτασίας ἱλάεσθαι). True, the motif of spitting three times also occurs in Theocritus, in a context that is likely to be reflected in the Ciris as well (6.39‒40 ὡς μὴ βασκανθῶ δέ, τρὶς εἰς ἐμὸν ἔπτυσα κόλπον· | ταῦτα γὰρ ἁ γραία με Κοτυτταρὶς ἐξεδίδαξε), since it is a closer parallel in that it speaks, like the Ciris but unlike Apollonius, of spitting into  Alluding to Od. . σοὶ [sc. for Agamemnon] δὲ Κλυταιμνήστρη δόλον ἤρτυε.

140

4 Heroic deeds (lines 349‒385)

the bosom. The Theocritean context is not, however, as harmless as it appears: Polyphemus spits to cheat the evil eye after seeing his own reflection in the sea, a portrait that with its details (the only eye: 6.36 ἁ μία κώρα, the white teeth: 37 τῶν δέ τ’ ὀδόντων | λευκοτέραν αὐγάν) evokes his encounter with Odysseus.²⁸ In view of this Homeric subtext, Polyphemus’ spitting in Theocritus could perhaps be linked to Homer’s gross depiction of his vomiting (Od. 9.373‒374 φάρυγος δ’ ἐξέσσυτο οἶνος | ψωμοί τ’ ἀνδρόμεοι· ὁ δ’ ἐρεύγετο οἰνοβαρείων) after having drunk three cups of wine (361 τρὶς μὲν ἔδωκα φέρων, τρὶς δ’ ἔκπιεν ἀφραδίῃσιν). Be that as it may, the Theocritean context seems also to be intertextually linked to the Apollonian one:²⁹ apart from the motif of spitting three times, we may point out that both share the verse-ending ὀδόντων. Now, the motif of spitting is, in fact, one that can be demonstrated to have attracted, because of its rare and peculiar use in Homer, special attention from Hellenistic and Roman poets. The unprefixed πτύειν is a Homeric hapax (Il. 23.697 αἷμα παχὺ πτύοντα), as is ἐκπτύειν (Od. 5.322‒323 στόματος δ’ ἐξέπτυσεν ἅλμην | πικρήν).³⁰ Apollonius simultaneously alludes to both contexts: he uses the prefixed verb (though in tmesis), in the same metrical position and in the same tense as in Homer, but the substance that is spat (blood) evokes rather the use of the unprefixed verb. An almost identical case of double allusion to the two Homeric hapaxes is Theocr. 22.98‒99 ἐκ δ’ ἔπτυσεν αἷμα | φοίνιον: the situation (boxing contest) clearly points to the Iliadic context, whereas formal features – the use of ἐκπτύειν rather than πτύειν, its metrical position, the adjective in enjambement (ἅλμην | πικρήν, cf. αἷμα | φοίνιον) – allude to the Odyssean one. In a similar vein, it has been argued that Virgil’s only use of spuere (Georg. 4.97) shows his awareness that πτύειν is a Homeric hapax.³¹ Furthermore, when Statius tells of Leucothea’s sorrow over the dead body of Palaemon vomiting seawater (Th. 9.403 saeuum mare respuit infans), he intends a pointed reference to the Odyssean context (note that respuit takes the same metrical position as [ἐξ]έπτυσεν in Homer), which only a few lines later speaks of Leucothea saving Odysseus from drowning (contrast also Od. 5.334 Λευκοθέη, ἣ πρὶν μὲν ἔην βροτός with Th. 9.401‒402 nondum Nereida … Leucothean): Leucothea helps Odysseus because she is reminded of her own son. It is therefore quite likely that the Ciris poet, especially given his interest in Homeric hapaxes (cf. below), would have taken account of more than just

 Cf. Hunter () : “any mention of the Cyclops’ teeth will evoke the use to which he was to put them in Odyssey .”  Whatever their chronological relationship.  In addition, ἀποπτύειν has three occurrences: Il. ., . and, as a variant reading, Od. ..  Farrell () ‒.

4.7

141

one poetic context of (‐)πτύειν (note, again, that the first despue is placed in the same metrical position as ἔπτυσ〈ε〉 in Apollonius and ἔπτυσα in Theocritus). I would therefore tentatively suggest that the triple ter may be intended to sum up, as it were, the two τρίς in Apollonius and the one in Theocritus.

4.7 The next two subtexts come closest to the basic formula ‘the hero kills the monster’, as they, in a sense, cast Scylla in the role of a dragon-slayer. The first relevant passage – though strictly speaking it refers to Carme rather than Scylla – describes the ritual of defixio aimed at subduing Nisus’ will (376‒377): pergit Amyclaeo spargens altaria thallo | regis Iolciacis animum defigere uotis. It may be interesting to note that, on the literal level, the target of the magical action is none other than Nisus’ animus; in other words, we have yet one more sign that the basic epic conflict is here transferred into the sphere of ‘intellectual’, rather than physical, confrontation. This passage, however, contains a remarkable allusion that sheds a completely new light on it. To begin with, thallus is a Greek loanword, and it occurs in poetry only here (and is very rare elsewhere in Latin), which, of course, invites the suspicion of Greek influence. Moreover, θαλλός is a Homeric hapax (Od. 17.224), and although not so rare after Homer, it has likewise only one occurrence in Apollonius (4.156‒159): ἡ δέ μιν ἀρκεύθοιο νέον τετμηότι θαλλῷ, βάπτουσ’ ἐκ κυκεῶνος, ἀκήρατα φάρμακ’ ἀοιδαῖς ῥαῖνε κατ’ ὀφθαλμῶν, περί τ’ ἀμφί τε νήριτος ὀδμὴ φαρμάκου ὕπνον ἔβαλλε.

This is a description of how Medea conquers the serpent’s vigilance when she and Jason come to take the Golden Fleece, and I would argue that it is the main model of the Ciris passage. In an earlier chapter we have seen another Homeric hapax, οἶστρος, being reused in the Ciris in such a way as to evoke its two contexts in Apollonius (2§10). Here we may observe that thallo is placed, like Apollonius’ θαλλῷ, at a verse-end (not the Homeric position), and also that uotis neatly catches the meaning of ἀοιδαῖς, both words being placed at the end of the following line. In fact, in the two passages we find the same combination of sprinkling and incantation; the only difference is perhaps that, whilst Medea uses magic on the serpent directly, Carme has to affect Nisus by means of sympathetic magic. Theocritus 24.98 θαλλῷ ἐπιρραίνειν ἐστεμμένῳ ἀβλαβὲς ὕδωρ seems to be the only other early context that speaks of the use of θαλλός

142

4 Heroic deeds (lines 349‒385)

for sprinkling, but in this case it is sprinkling with lustral water in a purification ritual, which is certainly not what Carme does in the Ciris. ³² Yet it is tempting to speculate, especially in view of its apparent link to the Apollonian context,³³ that the Theocritean context is also taken account of (for instance, it shares with the Ciris the motif of fuming with sulphur, 24.96 καθαρῷ δὲ πυρώσατε δῶμα θεείῳ, cf. 369 componens sulpura testa). If, then, the Ciris does here allude to the Argonautica, Nisus turns out to parallel the serpent guarding the Golden Fleece, whereas Scylla and Carme assume, as it were, the roles of Jason and Medea opposing the serpent. Of course, in the Argonautica the serpent is not killed, but merely put to sleep by Medea’s magic (in contrast, Carme’s magic proves unsuccessful); in other versions, however, Jason does kill the serpent, thus fulfilling what is expected from an Indo-European hero. I suggest that the Ciris alludes to such a version, as well as to the Argonautica. At the end of the present episode, in the passage introducing the next one, the reader faces an exceptional acceleration of the narrative pace (386‒390): ergo iterum capiti Scylla est inimica paterno: tum coma Sidonio florens deciditur ostro, tum capitur Megara et diuum responsa probantur, tum suspensa nouo ritu de nauibus altis per mare caeruleum trahitur Niseia uirgo.

As Lyne aptly comments: The narrative is grossly unbalanced, in the manner of epyllion […]. We should note that the poet elected to elaborate the failed attempt to cut the hair (206 ff.) rather than the cutting itself, and that there is nothing at all on the confrontation of Minos and Scylla after the event. This is indeed idiosyncratic emphasis.³⁴

This passage, however, has a striking parallel – I suspect, a direct model – in Pindar’s fourth Pythian ode (247‒251):³⁵

 Though one can agree with Lyne ()  that the wording in the Ciris is not explicit enough.  As White ()  suggests, the fact that Apollonius and Theocritus have θαλλῷ at the opposite ends of hexameter may be significant. I would also point out that in Theocritus the purification ritual is connected with snakes, which in a sense parallel Apollonius’ serpent.  Lyne () . Cf. Bartels () .  In addition to the two quoted passages, Nisus’ confidence ( tanta est in paruo fiducia crine cauendi) can be compared with that of Aeetes ( ἔλπετο δ’ οὐκέτι οἱ κεῖνόν γε πράξασθαι

4.7

143

μακρά μοι νεῖσθαι κατ’ ἀμαξιτόν· ὥρα γὰρ συνάπτει· καί τινα οἶμον ἴσαμι βραχύν· πολλοῖσι δ’ ἅγημαι σοφίας ἑτέροις. κτεῖνε μὲν γλαυκῶπα τέχναις ποικιλόνωτον ὄφιν, ὦ ᾿Aρκεσίλα, κλέψεν τε Μήδειαν σὺν αὐτᾷ, τὰν Πελίαο φονόν· ἔν τ’ Ὠκεανοῦ πελάγεσσι μίγεν πόντῳ τ’ ἐρυθρῷ…

We shall return again to this parallel in the next chapter (5§2), in order to discuss its poetological significance; at the moment I confine my remarks to the narrative dimension. Exactly like the Ciris, Pindar interrupts the regular flow of his account at the point when he has come to the central event of the story, obtaining the Golden Fleece; moreover, not only does he drastically speed up the narrative, but in fact he omits to mention the Fleece at all, much in the same way as the Ciris withholds the very fact of Nisus’ purple lock being handed over to Minos. On the formal level, the emphatic tricolon of the Ciris (note the triple tum) is paralleled, both as a whole and in its individual members, in Pindar: the cutting of the hair is analogous to the killing of the serpent, the sack of Megara to the abduction of Medea, the return sea voyage of Minos (joined by Scylla) to that of the Argonauts. Furthermore, the colour contrast of the purple lock (ostro) and the green sea (caeruleum) chiastically reproduces that of the green-eyed serpent (γλαυκῶπα) and the Red Sea (ἐρυθρῷ). Also, the use of the passive voice throughout the Ciris passage (deciditur, capitur, probantur, trahitur), without any indication of the actual agent, neatly parallels the conspicuous absence of an explicit subject with Pindar’s active verbs (κτεῖνε, κλέψεν, μίγεν: the last is passive only morphologically). Watkins has revealing ideas to offer on the Pindar passage, which may be useful to quote at some length: Pindar claims to be in a hurry and wants to abridge his narrative. He says, ‘I know a shortcut’ and then, ‘I am a leader in the lore of song’, i. e., ‘a master of poetic tradition’. The next line proves his assertion, for the shortcut is the traditional basic formula itself. […] Encapsulating the myth of Jason and the Golden Fleece – without overt subject –, Pindar begins the sentence and the verse line with the verb κτεῖνε ‘slew’, ends the verse-line with the object ὄφιν ‘serpent’, and then ends the sentence and the next verse-line with a form of the root *gṷhen-, φονός ‘murderess’. […] [It] reintroduces the marked root for ‘kill’ *gṷhen-, vis-àvis the common, unmarked κτείνω. By a process of semantic spreading it has given to κτεῖνε (in emphatic line-initial position) the semantics of πέφνε; with the object ὄφιν (distracted to line-final focus position) the effect is to restore the Indo-European basic formula.³⁶

πόνον): Nisus has faith in the magic hair much in the same way as Aeetes has faith in the serpent.  Watkins () .

144

4 Heroic deeds (lines 349‒385)

As noted above, the cutting of Nisus’ hair (deciditur) is intertextually linked to the killing of the serpent (κτεῖνε); now I would suggest that this parallel implies more than just a structural analogy. The Latin verb is not the most obvious choice for the context, as one might rather expect a form of tondere, the terminus technicus for cutting hair which plays on the etymology of ciris. Although the prefixed decidere may sound harmless enough, it seems to be put here so as to recall the unprefixed caedere, used a few lines earlier (366 caesa pio cecidisset uictima ferro) in the sense of ‘to kill, slay, murder’ (OLD, s.v., 3). So we may perhaps conclude that Scylla’s act of cutting Nisus’ hair is consciously intended by the Ciris poet as a variation on the theme ‘the hero kills the monster’.

5 An epic voyage (lines 386‒477) 5.1 The account of Scylla’s sea voyage behind the stern of Minos’ ship (386‒477), during which she delivers her last monologue (404‒458), seems both straightforward and puzzling at the same time. What puzzles is, in fact, the very straightforwardness with which this lengthy speech – “uttered with surprising fluency considering Scylla’s uncomfortable position”¹ – is inserted in the otherwise concise, almost technical, narrative. Neither Minos, nor any other member of the crew appears to be present; the impression is that Scylla sails entirely on her own (Minos is only mentioned in Scylla’s speech, but not in the frame narrative). Thus Scylla’s monologue turns out to be, as it were, the sole purport of the voyage, and in this identification of the two rather incongruent activities we can easily recognise a traditional poetological metaphor: song as ship (or sea voyage). The metaphor, Indo-European in origin,² is well attested in early Greek poetry³ and, later, is recurrently evoked in Latin poetry.⁴ Perhaps most interestingly, it has been argued to be a constituent trope for Apollonius, in whose Argonautica the progress of the expedition is in many ways implicitly identified with the act of narration.⁵ In the Ciris, I would suggest, this metaphor is already adumbrated in the proem, where a didactic epic is figuratively spoken of as the Panathenaic peplos which traditionally was carried as a sail of a wheeled ship⁶ (21‒41), whilst here, as we shall see, it is developed with a much greater thoroughness in both the frame narrative and the monologue itself.

5.2 As was suggested in the preceding chapter, the passage opening the sea-voyage episode of the Ciris has a model in an explicitly poetological context in Pindar’s fourth Pythian ode. Although the metaphor employed there is not exactly the  Lyne () .  West () ‒.  Nünlist () ‒.  Harrison ().  Albis ().  Which is modelled on the Argo in Catullus : cf. Ciris ‒ cum leuis alterno Zephyrus concrebuit Euro | et prono grauidum prouexit pondere currum with . ipsa leui fecit uolitantem flamine currum.

146

5 An epic voyage (lines 386‒477)

same as the one I argue to be at work in the Ciris (‘the path of song’, οἶμος, rather than ‘the sea voyage’), such a model obviously invites one to look for a poetological content in the imitation as well. To start with an indirect argument, we may note an Apollonian allusion to this metaphorical context of Pindar’s ode (Arg. 4.838‒841): ἀλλ’ ὥρη δολιχήν τε καὶ ἄσπετον οἶμον ὁδεύειν, ὄφρα κασιγνήτας μετελεύσομαι αἵ μοι ἀρωγοὶ ἔσσονται, καὶ νηὸς ὅθι πρυμνήσι’ ἀνῆπται, ὥς κεν ὑπηῷοι μνησαίατο νόστον ἑλέσθαι.

As Albis demonstrates on internal grounds, this is an “instance of the narrator describing his characters’ activity with vocabulary equally applicable to his own poetic activity,” for, as he argues, the terms οἶμος and νόστος are heavily loaded, here and elsewhere in Apollonius, with poetological connotations.⁷ I would add that Thetis’ words – ἀλλ’ ὥρη δολιχήν τε καὶ ἄσπετον οἶμον ὁδεύειν – reflect, with a touch of humour, those of the Pindaric narrator (Pyth. 4.247‒248): μακρά μοι νεῖσθαι κατ’ ἀμαξιτόν· ὥρα γὰρ συνάπτει· καί τινα | οἶμον ἴσαμι βραχύν.⁸ On the one hand, Thetis uses time constraints as an excuse to break off her conversation with Hera, much in the same way as Pindar does to speed up the narrative. On the other hand, and this is where the poetological implication becomes apparent, in Apollonius, just as in Pindar, it is a passage that (re‐)announces the Argonauts’ return journey. There may be a point in that it is not any sea voyage, but specifically a homebound voyage, that is poetologically significant: this apparently is a reflection of the epic tradition of νόστοι. Scylla’s sea voyage, too, is – certainly from Minos’ perspective – a homeward journey.⁹

 Albis () ‒; note also: “Just as the Argonauts assume the role of the poet in recalling their nostos, Thetis assumes the role of Muse in prompting their act of recollection.” Speaking broadly, this is an example of the ‘poeta creator’ figure; we shall encounter a similar case in the Ciris later on.  Note the pointed contrast between Apollonius’ ‘I have no time because I have a long way to go’ and Pindar’s ‘I shall not go the longer path because I have no time’.  But also from Carme’s: ‒ reuehi quod moenia Cretae | gaudeat. Note Virgil’s only use of the verb in Aen. .‒ Troianam ex hostibus urbem | qui reuehis nobis (Aeneas’ journey projected against the tradition of νόστοι as well as a reference to Dardanus’ provenance) and two Horatian contexts: Epod. . nec mater domum caerula te reuehet (of Achilles) and S. .. non satis est Ithacam reuehi (of Odysseus).

5.2

147

Another indirect confirmation can be derived from what arguably is an allusion to the Ciris context, Virgil’s account of Aeneas’ river voyage to the site of the future Rome (Aen. 8.90‒92):¹⁰ ergo iter inceptum celerant rumore secundo: labitur uncta uadis abies; mirantur et undae, miratur nemus insuetum fulgentia longe scuta uirum fluuio pictasque innare carinas.

Most conspicuously, the repetition of mirari evokes the same figure in Ciris 391‒ 392 complures illam nymphae mirantur in undis, | miratur pater Oceanus. ¹¹ Soon we shall return to the motif of wonder and discuss its poetological connotations in greater detail. More relevantly at the moment, we can note an echo of the line introducing the sea-voyage episode (386): ergo iterum capiti Scylla est inimica paterno. To start with, Virgil seems to be playing on the derivation of iterum from iter. ¹² This point of contact, at first sight unremarkable if not accidental, emphasises a theme of profound importance: the theme of resuming an enterprise, which also implies resuming the narration of this enterprise. In the Ciris the adverb iterum refers to Scylla’s resuming her interrupted attempt to cut off Nisus’ purple lock; in the Aeneid Aeneas resumes a delayed journey (iter inceptum, pointing back to 8.80 socios simul instruit armis: cf. Ciris 381 rursus ad inceptum sociam se adiungit alumnae). As is demonstrated by Nelis, Aeneas’ river voyage is modelled on a number of poetologically charged passages describing the Argo’s sea voyage,¹³ some of which, as we shall see, also constitute a subtext for the present section of the Ciris. To make a more specific point, we may speculate that, in particular, Virgil’s iter has poetological implications similar to those of Apollonius’ (and Pindar’s) οἶμος, for indeed Aeneas resumes his journey as Virgil resumes his account of it.¹⁴ Although iter is a term not explicitly present in the Ciris, the idea of resuming an interrupted narrative – after a lengthy (and, as it turns out, unnecessary) digression – is certainly there.

 Note that it is in this context that Virgil speaks of the ‘return’ of the Trojans to Italy (see the preceding footnote).  Contrast Munari () .  On the etymology, cf. Maltby () . In fact, the pun is perhaps even more complex as inceptum may be etymologising caput: on the derivation of caput from capere (initium), cf. Maltby () .  Nelis () ‒.  Note that ‘shouts of applause’ (rumore secundo) are a reaction as suitable (if not more) to a poetic performance as to the departure of a ship.

148

5 An epic voyage (lines 386‒477)

5.3 The most immediate model for the motif of wonder can be found, of course, in the prologue of Catullus 64 (14‒15): emersere feri candenti e gurgite 〈ponti〉 ¹⁵ | aequoreae monstrum Nereides admirantes. The striking triple anaphora of tum at 64.19‒21, imitated in Ciris 387‒389, is as good a formal confirmation as one may expect. In turn, the Catullan context is well known to be alluding to Apollonius’ account of the Argo’s first sailing (1.547‒552):¹⁶ πάντες δ’ οὐρανόθεν λεῦσσον θεοὶ ἤματι κείνῳ νῆα καὶ ἡμιθέων ἀνδρῶν γένος, οἳ τότ’ ἄριστοι πόντον ἐπιπλώεσκον. ἐπ’ ἀκροτάτῃσι δὲ νύμφαι Πηλιάδες σκοπιῇσιν ἐθάμβεον εἰσορόωσαι ἔργον ᾿Aθηναίης Ἰτωνίδος ἠδὲ καὶ αὐτοὺς ἥρωας χείρεσσιν ἐπικραδάοντας ἐρετμά.

In some points the Ciris comes closer to the Apollonian passage than Catullus does, which may suggest a direct connection: compare in particular nymphae mirantur with νύμφαι … ἐθάμβεον (in contrast to Catullus’ Nereides, derived from Arg. 4.930). As has been observed by Albis, the broader Apollonian context is replete with details suggesting that the Argo’s sea voyage can in a certain way be identified with the poem itself.¹⁷ Only a few lines before the quoted passage, the Argonauts rowing to Orpheus’ accompaniment are compared with a chorus performing a dance in honour of Apollo (1.536‒541). Likewise, a few lines after that passage, the Argo’s progress with Orpheus singing on board, followed by fish big and small, is compared with the shepherd leading a flock by the sound of his flute (1.573‒579): ἰχθύες ἀίσσοντες ὑπὲξ ἁλός, ἄμμιγα παύροις ἄπλετοι, ὑγρὰ κέλευθα διασκαίροντες ἕποντο. ὡς δ’ ὁπότ’ ἀγραύλοιο μετ’ ἴχνια σημαντῆρος μυρία μῆλ’ ἐφέπονται ἄδην κεκορημένα ποίης εἰς αὖλιν, ὁ δέ τ’ εἶσι πάρος, σύριγγι λιγείῃ καλὰ μελιζόμενος νόμιον μέλος· ὣς ἄρα τοί γε ὡμάρτευν· τὴν δ’ αἰὲν ἐπασσύτερος φέρεν οὖρος.

 Trappes-Lomax () ‒ seems right in reading feri … ponti rather than freti … uultus.  See e. g. Clare () ‒, Trimble ()  and ‒.  Albis () ‒.

5.3

149

Now, it seems natural to imagine – a point missed by Albis – that the nymphs’ astonishment at the sight of the Argo is not only meant to be taken literally, but can also imply the audience’s reaction to a novel, outstanding performance, a reaction not unlike that of the Argonauts to Orpheus’ cosmological song (1.513‒514): τοὶ δ’ ἄμοτον λήξαντος ἔτι προύχοντο κάρηνα | πάντες ὁμῶς ὀρθοῖσιν ἐπ’ οὔασιν ἠρεμέοντες. The close connection between the Argo’s sailing and Orpheus’ singing is acknowledged by Moschus, who presents Europa’s sea voyage as, almost, an actual musical procession (118‒124): Νηρεΐδες δ’ ἀνέδυσαν ὑπὲξ ἁλός, αἳ δ’ ἄρα πᾶσαι κητείοις νώτοισιν ἐφήμεναι ἐστιχόωντο. καὶ δ’ αὐτὸς βαρύδουπος ὑπεὶρ ἅλα Ἐννοσίγαιος κῦμα κατιθύνων ἁλίης ἡγεῖτο κελεύθου αὐτοκασιγνήτῳ· τοὶ δ’ἀμφί μιν ἠγερέθοντο Τρίτωνες, πόντοιο βαρύθροοι αὐλητῆρες, κόχλοισιν ταναοῖς γάμιον μέλος ἠπύοντες.

This is a clear case of imitatio cum variatione: νόμιον μέλος becomes γάμιον μέλος (though remaining in the same metrical position), ὑγρὰ κέλευθα … ἕποντο is mirrored with ἁλίης ἡγεῖτο κελεύθου (but cf. 152 ὑγρὰ κέλευθα), and ἀγραύλοιο … σημαντῆρος is apparently punned upon with αὐλητῆρες (note that both produce a spondaic ending). It is to this accompaniment that Europa utters her speech (135‒152), which in a sense is the closest comparandum to Scylla’s monologue, as it likewise coincides temporally with the act of sailing. It seems worth noting that Europa is explicitly likened to a ship (129‒130 κολπώθη δ’ ὤμοισι πέπλος βαθὺς Εὐρωπείης | ἱστίον οἷά τε νηός), a detail that may be meant to evoke the metaphor of ‘the ship of song’. When we come to discuss Scylla’s monologue, we shall find further and more specific points of contact between Moschus’ Europa and the Ciris; for now I confine myself to noting one minor parallel: both Scylla and Europa are expressly said to sail in an unprecedented manner (389 suspensa nouo ritu de nauibus, 148 ξείνην ναυτιλίην ἐφέπω).¹⁸ Thus I would suggest that the parade of sea deities gazing in wonder at Scylla (391‒399) has a poetological implication, too, for they are actually the audience of Scylla’s performance. Such an interpretation can be further supported, and elaborated, if we consider a second model of the Catullan passage we started with, namely Apollonius’ account of the Argo’s passage through the Wander-

 Which evokes the concept of the first ship, but also that of the novelty of a song.

150

5 An epic voyage (lines 386‒477)

ing Rocks. There as well we find gods watching the Argo, though in fear rather than astonishment (4.956‒960). But there is also another group of deities present there, Thetis with her sisters, who are actively involved in aiding the Argo’s progress. If the former can be interpreted as the audience, the latter assume the role of the performers, or the Muses, for they in fact determine the course of the narrative. Not only is their assistance crucial for the Argo’s successful passage through the Wandering Rocks, but moreover Thetis is responsible (on Hera’s orders) for the Argo’s taking the route through the Wandering Rocks rather than that between Scylla and Charybdis (4.825‒832).¹⁹ A similar case of divine intervention can be found in the Ciris, as Amphitrite takes pity on Scylla and transforms her into a bird²⁰ (484‒488): sed tamen aeternum squamis uestire puellam infidosque inter teneram committere pisces non statuit (nimium est auidum pecus Amphitrites): aeriis potius sublimem sustulit alis, esset ut in terris facti de nomine ciris.

We shall return to this passage in the next chapter (6§3), but one important observation can be made now: Amphitrite’s decision to transform Scylla into a bird rather than a fish is phrased in such a way as to recall the narrator’s choice of a particular version of the Scylla story announced in the proem (90‒91): potius liceat notescere cirin | atque unam ex multis Scyllam non esse puellis. From a poetological perspective, a deus ex machina, who can radically change the course of the story in whatever direction, is an obvious author figure.

 Which is a necessary choice, since the Argo’s passage through the Planctae is famously mentioned in Homer (Od. .‒), but at the same time a questionable one, because in Homer Odysseus is specifically instructed to take the other route as the less dangerous.  Amphitrite’s action (‒ non tulit ac miseros mutauit uirginis artus | caeruleo pollens coniunx Neptunia regno and ‒ sustulit alis, | esset ut in terris facti de nomine ciris) is reminiscent of Aphrodite’s rescuing Butes from the Sirens in Apollonius (.‒ ἀλλά μιν οἰκτείρασα θεὰ Ἔρυκος μεδέουσα | Κύπρις ἔτ’ ἐν δίναις ἀνερέψατο, καί ῥ’ ἐσάωσε, | πρόφρων ἀντομένη, Λιλυβηίδα ναιέμεν ἄκρην). Ovid’s version has further parallels with Apollonius, cf. Met. .‒ insilit undis | consequiturque rates faciente cupidine uires and .‒ ἔνθορε πόντῳ … θυμὸν ἰανθείς· | νῆχε δὲ πορφυρέοιο δι’ οἴδματος.

5.4

151

5.4 Perhaps the most unambiguous indication that Scylla’s sea voyage has a poetological dimension comes from a pointed internal allusion, as 391 complures illam nymphae mirantur in undis recalls a programmatic passage in the proem (54‒57): complures illam magni, Messalla, poetae (nam uerum fateamur: amat Polyhymnia uerum) longe alia perhibent mutatam membra figura Scyllaeum monstro saxum infestasse uoraci.

To start with, we may note that in both cases illam refers to Scylla: in the proem Scylla is said to have been the subject of many poems, now she attracts the attention of many nymphs. The link is strengthened by the fact that complures is a word, though very common in prose, noticeably avoided in poetry.²¹ It is a remarkable coincidence, if coincidence it be, that the only Horatian use of the word actually refers to the poets and critics considered by Horace competent judges of his poetry, among whom – another shared detail – Messalla is addressed (S. 1.10.84‒88):²² te dicere possum, Pollio, te, Messalla, tuo cum fratre, simulque uos, Bibule et Serui, simul his te, candide Furni, compluris alios, doctos ego quos et amicos prudens praetereo.

But most importantly, the passage in the Ciris proem is specifically alluding to Callimachus’ hymn to Delos, where in a similar manner the narrator is confronted by the need to choose a particular story to tell (4.28‒29): εἰ δὲ λίην πολέες σε περιτροχόωσιν ἀοιδαί, | ποίῃ ἐνιπλέξω σε; τί τοι θυμῆρες ἀκοῦσαι; We already have discussed what implications this allusion has for the proem (1§8), whilst here we may observe how it is resumed in the account of Scylla’s sea voyage. First of all, the line mentioning Oceanus and Tethys among the divine spectators

 Cf. Lyne () .  In addition, S. .. inconposito dixi pede currere uersus and ‒ num rerum dura negarit | uersiculos natura magis factos et euntis | mollius, ac siquis pedibus quid claudere senis, | hoc tantum contentus, amet scripsisse ducentos | ante cibum uersus may be alluding to Ciris  molli libeat pede claudere uersum and  naturae rerum magnis intexere chartis.

152

5 An epic voyage (lines 386‒477)

(392 miratur pater Oceanus et candida Tethys) is clearly modelled on a line in Callimachus’ description of the procession of the islands led by Delos (4.16‒18):²³ ἀλλά οἱ οὐ νεμεσητὸν ἐνὶ πρώτῃσι λέγεσθαι, ὁππότ’ ἐς Ὠκεανόν τε καὶ ἐς Τιτηνίδα Τηθὺν νῆσοι ἀολλίζονται, ἀεὶ δ’ ἔξαρχος ὁδεύει.

As an additional confirmation of the link between the Ciris and Callimachus’ hymn, I would point out that the same two Callimachean passages are imitated, within a single episode, by Ovid (F. 5.5‒6 sic, quia posse datur diuersas reddere causas, | qua ferar ignoro, copiaque ipsa nocet and 81 duxerat Oceanus quondam Titanida Tethyn), in a context that also parallels the Ciris in mentioning Polyhymnia (5.9 and 53).²⁴ It is of crucial importance that the procession of the islands is conceived of as a musical procession, ἔξαρχος being “terminus technicus for the leader of a choir, a thiasos, etc.”²⁵ This motif is further developed when Delos is presented as the centre of a chorus of dancing islands (4.300‒301 σὲ μὲν περί τ’ ἀμφί τε νῆσοι | κύκλον ἐποιήσαντο καὶ ὡς χορὸν ἀμφεβάλοντο). In fact, as Slings has argued in detail, in this hymn Delos can be taken as an allegorical representation of Callimachus’ own poetry.²⁶ Scylla, however, is associated with the Delos of Callimachus’ hymn not only at this moment when she is, as it were, applauded by Oceanus and Tethys, but also in two further contexts. Let us begin with an allusion to the passage that immediately precedes the mention of the procession of the islands, the first description of Delos in the hymn (4.11‒15): κείνη δ’ ἠνεμόεσσα καὶ ἄτροπος οἷά θ’ ἁλιπλὴξ αἰθυίῃς καὶ μᾶλλον ἐπίδρομος ἠέπερ ἵπποις πόντῳ ἐνεστήρικται· ὁ δ’ ἀμφί ἑ πουλὺς ἑλίσσων Ἰκαρίου πολλὴν ἀπομάσσεται ὕδατος ἄχνην· τῷ σφε καὶ ἰχθυβολῆες ἁλίπλοοι ἐννάσσαντο.

This is a significant passage, and it has often been suspected that its water imagery conveys a poetological message; as, for example, Slings comments: πολλὴν ἀπομάσσεται ὕδατος ἄχνην may be taken as a direct reply to pthonus’ words: οὐκ ἄγαμαι τὸν ἀοιδὸν ὃς οὐδ’ ὅσα πόντος ἀείδει (Hy. 2.106) […] The foam that comes in great quantities from the πόντος […] might be taken as opposed to the quantity that pthonus re-

   

Cf. Dal Zotto () . Elsewhere in poetry she is only named by Horace (Carm. ..) and Martial (..). Mineur () . Moreover, it is a Homeric hapax (Il. .). Slings ().

5.4

153

quires: the πόντος, πουλὺς ἑλίσσων, leaves only the very thinnest trace of all its waters on Delos, the symbol of Callimachus’ poetry.²⁷

This passage has also notably been argued to lie behind Lucretius’ description of Sicily (1.718‒719): quam fluitans circum magnis anfractibus aequor | Ionium glaucis aspargit uirus ab undis. ²⁸ I would further suggest that it is likewise alluded to in the Ciris, when Scylla’s metamorphosis into the eponymous bird has finally taken place (514‒519): quae simul ut sese cano de gurgite uelox cum sonitu ad caelum stridentibus extulit alis et multum late dispersit in aequora rorem, infelix uirgo nequiquam a morte recepta incultum solis in rupibus exigit aeuum, rupibus et scopulis et litoribus desertis.

To start with, the fourth-foot homodyne dispersit neatly captures (much as Lucretius’ aspargit does) Callimachus’ similarly positioned ἀπομάσσεται. Perhaps the most precise and remarkable parallel is that between πολλὴν ἄχνην and multum rorem. Anglophone scholars tend to translate ἄχνη in that and similar contexts as ‘foam’ without specification,²⁹ but this translation is misleading. As is explained in LfgrE (s.v., 2), ἄχνη is Gischt, Schaum, wohl nicht zufällig, anders als beim fast synon. ἀφρός (s.d.), nur vom Schaum auf Meereswellen, u. zwar nur speziell von der hoch spritzenden Gischt, die sich beim Aufprall der Wellen auf ein Hindernis (Steilküste, Schiff) oder bei starkem Sturm bildet.

The Homeric expression ἁλὸς ἄχνη (Il. 4.426, Od. 6.403), which Callimachus only slightly varies, must therefore be understood exactly as ‘sea-water spray’, for which ros is a close equivalent (OLD, s.v., 2a).³⁰ In addition to this specific textual parallel, it can be pointed out that Scylla’s living, when transformed into a bird, solis in rupibus and litoribus desertis may hint at Callimachus’ portrayal of Delos

 Slings () .  Brown () .  See e. g. Stephens ()  and .  Ciris  multum late dispersit in aequora rorem is often assumed to be modelled on Verg. Georg. . rorem late dispergit amarum, which in turn evokes Od. . πικρὸν ἀποπνείουσαι ἁλὸς πολυβενθέος ὀδμήν; but as we just have seen, rorem is more likely to have originated as translation of Callimachus’ ἄχνην than of Homer’s ὀδμήν.

154

5 An epic voyage (lines 386‒477)

as a lonesome habitat of shearwaters and, later in the hymn, seals (4.242‒243 ὅθι φῶκαι | εἰνάλιαι τίκτουσιν, ἐνὶ σπιλάδεσσιν ἐρήμοις).³¹ The third and last point of contact between Scylla and the Callimachean Delos is more subtle and complex. After Scylla’s monologue there follows a detailed account of her itinerary, from the Isthmus to the Cyclades (463‒477), the geographical anomalies of which have often been taken as a sign of the Ciris poet’s gross incompetence.³² As Lyne observes, “[a]lthough there are decided oddities in 463‒72 (the journey within the Saronic Gulf), the real surprises occur in the peregrination around the Cyclades in 473‒7”³³ (471‒475): florentisque uidet iam Cycladas: hinc Venus illi Sunias, hinc statio contra patet Hermionaea. linquitur ante alias longe gratissima Delos Nereidum matri et Neptuno Aegaeo; prospicit incinctam spumanti litore Cythnum.

On leaving the Saronic gulf, Scylla finds herself between two protruding pieces of land, that of Attica on the one side and Argolis on the other; at this point she has her first glimpse of the Cyclades (uidet iam Cycladas). Straight after that, and this is indeed the greatest surprise, Scylla is said to leave behind (linquitur) Delos, which of course should be right in the middle of the Cyclades, whereas Cythnus, the first island of the group to be met when one sails from the Saronic gulf, is still in front of her (prospicit). I suggest that this is a learned inaccuracy.³⁴ Gall has tried to eliminate the anomaly by changing Delos to Cea, the name of the furthest northern island of the Cyclades³⁵ (though it seems, in fact, to be slightly out of the route). However, exactly at the point where the harbour of Hermione, no more hidden by Scyllaeum, comes into view (statio contra patet Hermionaea), there is a small island (Belbina, modern San Giorgio) that for its size and shape can be confused with Delos. It can be speculated that such a confusion may have been treated in literary sources known to the Ciris poet; and, in fact, we actually encounter Delos within the Saronic gulf in Callimachus’ hymn (4.41‒50): πολλάκι σε Τροιζῆνος ἀπὸ †ξάνθοιο† πολίχνης ἐρχόμενοι Ἐφύρηνδε Σαρωνικοῦ ἔνδοθι κόλπου

    

Cf. Dal Zotto () . See Lyne () ‒. Lyne () . It should also be kept in mind that the text is badly transmitted. On deliberate inconsistencies in Roman epic see in general O’Hara (). Gall () .

5.4

155

ναῦται ἐπεσκέψαντο, καὶ ἐξ Ἐφύρης ἀνιόντες οἱ μὲν ἔτ’ οὐκ ἴδον αὖθι, σὺ δὲ στεινοῖο παρ’ ὀξὺν ἔδραμες Εὐρίποιο πόρον καναχηδὰ ῥέοντος, Χαλκιδικῆς δ’ αὐτῆμαρ ἀνηναμένη ἁλὸς ὕδωρ μέσφ’ ἐς ᾿Aθηναίων προσενήξαο Σούνιον ἄκρον ἢ Χίον ἢ νήσοιο διάβροχον ὕδατι μαστὸν Παρθενίης (οὔπω γὰρ ἔην Σάμος), ἧχί σε νύμφαι γείτονες ᾿Aγκαίου Μυκαλησσίδες ἐξείνισσαν.

It is a remarkable detail that Callimachus tells of Delos’ wanderings not in abstract terms, but by saying, against all chronological probability, that the island was often seen in the Saronic gulf by sailors (the invention of navigation cannot possibly predate Apollo’s birth): this, too, may be a pointer at a tradition that actually claimed that Delos, or an island like Delos, had occasionally been seen in the Saronic gulf. Whether or not such a tradition existed, it seems likely that the Ciris alludes to this Callimachean context. Scylla’s sea voyage takes place roughly in the same waters where Delos drifted: both texts explicitly mention Corinth (464 Corinthum, Ἐφύρηνδε), Athens and Sunium (Venus … Sunias, 469 Athenas, ᾿Aθηναίων … Σούνιον ἄκρον), and finally the Cyclades (uidet iam Cycladas, 198 Κυκλάδας ὀψομένη). The puzzling reference to Delos as ante alias longe gratissima Delos | Nereidum matri et Neptuno Aegaeo (paralleled in Verg. Aen. 3.73‒74) can be linked to Delos’ famous claim (4.268‒271): αὕτη ἐγὼ τοιήδε· δυσήροτος, ἀλλ’ ἀπ’ ἐμεῖο Δήλιος ᾿Aπόλλων κεκλήσεται, οὐδέ τις ἄλλη γαιάων τοσσόνδε θεῷ πεφιλήσεται ἄλλῳ, οὐ Κερχνὶς κρείοντι Ποσειδάωνι Λεχαίῳ…

The conflation of Delos with Corinth (Κερχνίς) as Neptune’s favourite place would have been, of course, deliberate.³⁶ However, the position of Delos is, though the most striking, not the only geographical anomaly of Scylla’s itinerary, as some of the other Cyclades, too, appear misplaced. But Callimachus’ hymn may provide an explanation for this as well. Although Delos’ mobility is a feature that distinguishes it from other islands (4.30‒36), the latter still are quite capable of fleeing at the approach of Leto: 4.158‒159 αἱ δ’ ὑπ’ ὀμοκλῆς | πασσυδίῃ φοβέοντο κατὰ ῥόον ἥντινα τέτμοι,

 Lyne ()  comments: “The effect is the odder because we must now [i. e. in the Ciris] take the superlative gratissima literally, emphasised indeed by ante alias longe, whereas Vergil’s was more loosely intensive.” This ante alias, omitted in Virgil, is a precise equivalent of οὐδέ τις ἄλλη. Cf. also . Κέρκυρα φιλοξεινωτάτη ἄλλων.

156

5 An epic voyage (lines 386‒477)

196 αἱ μὲν τόσσα λέγοντος ἀπέτρεχον εἰν ἁλὶ νῆσοι. It is hardly a coincidence that geographical inconsistency is specifically associated with the Cyclades in at least two further contexts in Latin poetry. One is Cydippe’s incongruous account of her visit to Delos in Ovid (Her. 21.81‒82).³⁷ The other is Virgil’s narration (3.124‒127), likewise inaccurate, of the Trojans’ sailing from Delos to Crete (which is also Minos’ destination in the Ciris).³⁸ As Barchiesi has amply demonstrated, the narrative of Aeneid 3 is indebted to Callimachus’ hymn to Delos;³⁹ of particular interest is his following observation: The navigation to Delos features the first occurrence in extant Latin poetry of the bold trope in which a landscape ‘runs away’ from the sight of a departing viewer: […] Before the safe landing in Delos mirrored in Aeneas’ narrative, Apollo and his mother went through a surrealistic experience: the places they approached were literally […] ‘on the run’ to avoid them: […] Leto is a persecuted runaway, and every possible goal runs away from her. It could be suggested that this experiment in the persecutory animation of landscapes has influenced the strong Vergilian imagery which represents Aeneas’ journey as the pursuit of a fleeing Italy by a restless exile.⁴⁰

I would, however, argue that in Latin poetry the motif of the moving landscape⁴¹ had already been associated with Callimachus before Virgil. Although the fact is not generally acknowledged, Catullus’ description of guests gathering at Peleus’ palace is clearly intended to evoke certain passages from Callimachus’ hymn (64.31‒37): quae simul optatae finito tempore luces aduenere, domum conuentu tota frequentat Thessalia, oppletur laetanti regia coetu; dona ferunt prae se, declarant gaudia uultu. deseritur Cieros; linquunt Pthiotica Tempe Crannonisque domos ac moenia Larisaea; Pharsalum coeunt, Pharsalia tecta frequentant.

Catullus begins with a personification-like metonymy (conuentu tota frequentat | Thessalia: contrast frequentant in 64.37), and then gradually his language be See Kyriakidis () ‒, though he offers a different explanation.  Cf. Horsfall () ‒. This context is certainly related to Ciris ‒. Cf. further Aen. .‒ pelago credas innare reuulsas | Cycladas, with P. Hardie’s ()  n.  discussion.  Barchiesi ().  Barchiesi () ‒.  For a treatment of the theme in Greek poetry (including Callimachus’ hymn to Delos), see Nishimura-Jensen (); cf. further Klooster (), focusing specifically on the hymn.

5.5

157

comes literal (linquunt Pthiotica Tempe | Crannonisque domos ac moenia Larisaea): this can be compared with Callimachus’ more or less naturalistic description of an earthquake (4.137‒138 ἔτρεμε δ’ Ὄσσης | οὔρεα καὶ πεδίον Κραννώνιον), though with a touch of personification (4.139‒140 φόβῳ δ’ ὠρχήσατο πᾶσα | Θεσσαλίη). When, however, Catullus comes to speak of the divine guests (64.278‒279 e uertice Peli | aduenit Chiron, 285‒287 confestim Peneos adest, uiridantia Tempe … linquens: note the parallelism with linquunt Pthiotica Tempe), his model is Callimachus’ account of the flight of the Thessalian localities personified (4.103‒105 ἂψ δ’ ἐπὶ Θεσσαλίην πόδας ἔτρεπε· φεῦγε δ’ Ἄναυρος | καὶ μεγάλη Λάρισα καὶ αἱ Χειρωνίδες ἄκραι, | φεῦγε δὲ καὶ Πηνειὸς ἑλισσόμενος διὰ Τεμπέων).

5.5 In the preceding sections I was arguing that the account of Scylla’s journey behind Minos’ ship is meant to evoke the metaphor of the sea voyage of song by alluding to two poetologically charged figures: the Apollonian (though not only Apollonian) Argo and the Callimachean Delos. It is now time to consider how Scylla’s lengthy soliloquy functions as part of that metaphor. Of course, the very contemporaneity of the two processes – sailing and speaking – is the most visible way in which the ‘sea voyage of song’ metaphor is manifested in the Ciris. However, it seems possible to point out more specific details that make one think of Scylla’s monologue as almost a poetic performance rather than just a desperate outburst. As argued in an earlier chapter (3§9), Carme’s longer speech features certain qualities of a formal piece of poetry and is thus intended to parallel, or perhaps rather prefigure, Scylla’s monologue, as both are explicitly defined as laments. Carme’s expertise as a poet, I suggested, can be seen, first of all, in her ability to draw analogies between situations that on the surface may appear disparate rather than similar, such as the diametrically opposed relations of Britomartis and Scylla to Minos. In fact, the two speeches themselves constitute a case of precisely this kind of (somewhat forced) analogy, for, although they are both termed questus (285, 401), and although both speakers complain about Minos, Carme’s is basically a funerary lament and Scylla’s an amatory one. Unlike Carme, Scylla does not draw explicit parallels in her monologue, but, as we shall see, it is replete with references to earlier literary contexts dealing with the theme of the abandoned heroine. Of course, the complex allusive tex-

158

5 An epic voyage (lines 386‒477)

ture of Scylla’s speech should not be taken literally to imply that she actually makes, as it were, these allusions herself.⁴² The identification of Scylla as a poet figure is more subtle, but it seems to be there. It is no coincidence that Scylla’s invocation of Procne at the beginning of her speech points back to a narratorial apostrophe earlier in the poem similarly addressed to the Dauliades (408‒ 411 and 198‒202): uos, uos, humana si qui de gente uenitis, cernitis? illa ego sum cognato sanguine uobis Scylla (quod o salua liceat te dicere, Procne), illa ego sum Nisi pollentis filia quondam. uosque adeo, humani mutatae corporis artus, uos o crudeli fatorum lege, puellae Dauliades, gaudete: uenit carissima uobis cognatos augens reges numerumque suorum ciris et ipse pater. uos, o pulcherrima quondam…

And a few lines later, Scylla ‘quotes’ it again: 423‒424 quorum direptis moenibus urbis | o ego crudelis flamma delubra petiui, cf. 191 cui direpta crudeliter urbe. This internal allusion, I suggest, is meant to signal that Scylla should be viewed as in a way equal with the narrator. Although strictly speaking that apostrophe is not placed at the beginning of the Ciris’ narrative part, it certainly can be said to introduce its central and dramatically most developed episode, that is to say the main action of the entire poem.⁴³ Accordingly, Scylla’s monologue, likewise opening with an invocation of the same mythological figures, can be read, from a certain perspective, as her own, ‘subjective’, version of the poem. There is also another argument that can be derived from Scylla’s address to Procne. As Lyne comments: By some accounts Tereus as hawk or hoopoe perpetually pursued Philomela and Procne in their swallow and nightingale shapes in desire for vengeance […]. Scylla’s reference to and concern for her cousin’s persecution are […] ominous and ironic, in that an exactly parallel fate (pursuit) fairly soon awaits herself.⁴⁴

 Though cf., on Catullus’ Ariadne, Trimble () : “Ariadne may have read other poems too.”  Also, although Scylla does not know yet what actually awaits her, she is able to foretell, as it were, her future fate:  non Libys Assyrio sternetur lectulus ostro, anticipating  non thalamus Syrio fragrans accepit amomo.  Lyne () .

5.6

159

If an analogy between Scylla transformed into the ciris and Procne transformed into the nightingale (rather than swallow) is indeed implied (note cognato sanguine uobis),⁴⁵ it is tempting to speculate that Scylla’s lament is in a way like that of the nightingale (ἀηδών), an archetypal poet figure (ἀοιδός).⁴⁶

5.6 As before, we may start by considering first the main Catullan model of Scylla’s monologue: Ariadne’s reproachful speech addressed to Theseus (64.132‒201).⁴⁷ Most generally, both belong to the class of speeches of betrayed heroines. It has been noted that against this Catullan background Scylla’s lament appears unorthodox;⁴⁸ and indeed, although the Ciris does have here a substantial number of unmistakable allusions to Catullus, it is the differences that are more important. Trimble offers a convenient structural analysis of the Catullan speech, with three main sub-divisions: 132‒63, reproaches addressed to Theseus; 164‒87, Ariadne’s lament about her current situation, with no specified addressee; 188‒201, her curse, invoking the Furies. Essentially 132‒63 concern the past on Crete, 164‒87 the present on Dia, and 188‒201 the future in Athens.⁴⁹

In Scylla’s monologue a comparable temporal structure can be observed, though the sequence is different: after the lengthy invocation of the winds, the birds, and finally Minos (404‒415), Scylla gives her own version of what happened (416‒437), then she speaks of the future, though in negative terms as she does not actually have any future (438‒447), and last she depicts her present situation, ending with yet another appeal to Minos (448‒458). This alteration is no doubt rhetorically justified: whereas in Catullus the climax falls on the curse directed at Theseus’ future,⁵⁰ Scylla’s lament, which does not have such a curse,

 And it should further be noted that in the sixth eclogue Virgil speaks of both Scylla and Philomela within a single passage (.‒); moreover, . sua tecta super uolitauerit alis, used of Philomela, is exactly identical to Ciris , used of Scylla.  On the etymological connection, cf. A. Hardie () .  On the latter see Trimble () ‒.  See Hross () .  Trimble () ; cf. Hross () .  Ariadne’s curse (.‒ sed quali solam Theseus me mente reliquit, | tali mente, deae, funestet seque suosque) has a remarkable model, not generally noted, in Theocritus (.‒ τόσσον ἔχοι λάθας ὅσσον ποκὰ Θησέα φαντὶ | ἐν Δίᾳ λασθῆμεν ἐυπλοκάμω ᾿Aριάδνας); though cf. Armstrong ()  n. .

160

5 An epic voyage (lines 386‒477)

culminates in what is capable of producing the greatest pathos, that is a vivid account of her desperate present situation. There may be different explanations of why Scylla does not curse Minos (some of which we shall look at later), but the fact remains that she does not.⁵¹ That this is no accidental omission, but rather a deliberate authorial choice, is confirmed by the pointed manner in which the Ciris alludes to the final section of Ariadne’s speech (64.188‒191): non tamen ante mihi languescent lumina morte, nec prius a fesso secedent corpore sensus, quam iustam a diuis exposcam prodita multam, caelestumque fidem postrema comprecer hora.

Scylla both begins and ends her monologue with a ‘quotation’ from this passage (405‒406 and 448‒450): dum queror et diuos (quamquam nil testibus illis profeci) extrema moriens tamen alloquor hora. iam fesso tandem fugiunt de corpore uires, et caput inflexa lentum ceruice recumbit, marmorea adductis liuescunt bracchia nodis.

Unlike Ariadne, Scylla explicitly claims that she expects no help from the gods (nil testibus illis | profeci), and this assertion is further substantiated at the level of allusion. Whereas Ariadne is sure that her last prayer will be fulfilled, Scylla realises the futility of her last appeal to the gods (note tamen); whereas Ariadne is certain she will not die before securing her objective, Scylla thinks she is about to die (contrast nec prius and iam) without any hope of being heard (though she makes her last attempt in 454‒458). Indeed, Scylla expects more compassion from the winds than from the gods (407 uos ego, uos adeo, uenti, testabor, et aurae, contrast 64.164 sed quid ego ignaris nequiquam conqueror auris?). This contrast is the more conspicuous as Scylla and Ariadne manifest a considerable mutual understanding in what they complain about. Each of them blames – Scylla Minos, Ariadne Theseus – for his deceitful good looks (431‒ 432 non equidem ex isto speraui corpore posse | tale malum nasci; forma uel sidera fallas, 64.175‒176 malus haec celans dulci crudelia forma | consilia). They both complain that they could not follow their beloved even as slaves (444‒446 mene alias inter famularum munere fungi … non licuit?, 64.160‒161 in uestras potuisti ducere sedes, | quae tibi iucundo famularer serua labore). They both fear  Cf. Bretzigheimer () ‒.

5.6

161

that after their imminent deaths they will not even receive proper funerals (441‒ 442 ne me … iniecta tellus tumulabit harena, 64.153 neque iniecta tumulabor mortua terra). As I have pointed out, the absence of a curse in Scylla’s monologue may have different explanations, which need not be mutually exclusive. One is that Scylla may be a different sort of person, a person completely controlled by just one emotion, love. We have already seen that, unlike the typical enamoured heroine, Scylla does not experience shame;⁵² likewise, we might think, she does not feel hatred. For not only does she not curse Minos, but neither does she use words like perfide or immemor about him; her reproaches are formulated in an extremely abstract or even impersonal way: 458 omnia nam potius quam te fecisse putabo, 427 iam iam scelus omnia uicit. There is, of course, also a purely pragmatic reason: in Catullus’ poem a curse is justified because the reader sees it fulfilled only a few lines later, whereas in the Scylla story as we know it no punishment awaits Minos. And a third explanation may lie in the use of other literary models by the Ciris poet: I shall formulate this later on. It is well known that Ariadne’s monologue has two main models: a speech by Medea in Euripides (Med. 465‒519) and two speeches, likewise by Medea, in Apollonius (4.355‒390 and 1031‒1052).⁵³ The speeches in Apollonius have evidently exercised a strong influence on Scylla’s lament, and the speech in Euripides, too, may have left some traces, although this is much less certain. Parallels with the Argonautica are concentrated in the opening section of Scylla’s monologue, in which she presents a view of her relation with Minos based on what passed between them. The main point in both Scylla’s and Medea’s speeches is that their past favours and sacrifices are not adequately rewarded. Both Scylla and Medea betrayed their fatherlands, homes and parents for the sake of either Minos or Jason (419‒420 quae sic patriam carosque penates | hostibus immitique addixi gnara tyranno, cf. 4.361‒362 πάτρην τε κλέα τε μεγάρων αὐτούς τε τοκῆας | νοσφισάμην and 1036‒1037 ἥδ’ ἐγώ, ἣ πάτρην τε καὶ οὓς ὤλεσσα τοκῆας, | ἣ δόμον). They both realise that what they did was a grave crime (424 o ego crudelis, cf. 4.360 ἐγὼ οὐ κατὰ κόσμον ἀναιδήτῳ ἰότητι and 380 δεινῶν ὕπερ οἷα ἔοργα), for which they can expect no pardon (418‒419 non equidem me alio possum contendere dignam | supplicio, cf. 4.1045‒1046 οὐκ ἀλεωρὴν | ἄλλην, οἰόθι δὲ προτιβάλλομαι ὑμέας αὐτούς). But they did it for love and in the hope that it would sanction their union with those they helped (414‒415 illa ego sum,

 Though, as Schmiel ()  rightly points out, Moschus’ Europa is likewise “quite unconcerned with αἰδώς or propriety”.  Hross ()  and ; cf. Trimble () .

162

5 An epic voyage (lines 386‒477)

Minos, sacrato foedere coniunx | dicta tibi, cf. 4.358‒359 ποῦ τοι Διὸς Ἱκεσίοιο | ὅρκια, ποῦ δὲ μελιχραὶ ὑποσχεσίαι βεβάασιν; and 368 τῶ φημὶ τεὴ κούρη τε δάμαρ τε). Instead, they have to suffer as they sail with their beloved (416 uinctane tam magni tranabo gurgitis undas? cf. 4.362‒363 τηλόθι δ’ οἴη | λυγρῇσιν κατὰ πόντον ἅμ’ ἀλκυόνεσσι φορεῦμαι); an immediate death would have been better (447 at belli saltem captiuam lege necasses, cf. 4.373‒374 ἢ σύγ’ ἔπειτα | φασγάνῳ αὐτίκα τόνδε μέσον διὰ λαιμὸν ἀμῆσαι). Once again, we must observe that despite all these close similarities Scylla does not parallel Medea in her insults (4.376 σχέτλιε, cf. 4.1047) and threats (4.386‒387 αὐτίκ’ ἐμαὶ ἐλάσειαν Ἐρινύες, cf. 4.1042). There seem to be no specific parallels with Medea’s speech in Euripides, with perhaps one exception. Unlike the Apollonian Medea who speaks of having lost her family (4.362 νοσφισάμην, 4.1036 ὤλεσσα), her Euripidean counterpart thinks of her actions in terms of betrayal (Med. 483‒485 αὐτὴ δὲ πατέρα καὶ δόμους προδοῦσ’ ἐμοὺς | τὴν Πηλιῶτιν εἰς Ἰωλκὸν ἱκόμην | σὺν σοί, πρόθυμος μᾶλλον ἢ σοφωτέρα, 502‒503 πατρὸς δόμους, | οὓς σοὶ προδοῦσα καὶ πάτραν ἀφικόμην). This closely resembles Scylla’s language (419‒420 patriam carosque penates | hostibus immitique addixi gnara tyranno, 428 tene ego plus patrio dilexi perdita regno?).⁵⁴ But there are in Euripides some possible points of contact outside that speech. To start with, it seems worth pointing out that Jason characterises Medea as Σκύλλης ἔχουσαν ἀγριωτέραν φύσιν (1343). Perhaps the most remarkable detail is Scylla’s claim that she would even have been happy to be Minos’ slave and serve his future wife (445‒446 coniugis atque tuae, quaecumque erit illa, beatae | non licuit grauidos penso deuoluere fusos?), which evokes Medea’s situation, in general, with Jason taking a new wife, and her affected goodwill as she sends her poisoned gifts, in particular (956‒958 λάζυσθε φερνὰς τάσδε, παῖδες, ἐς χέρας | καὶ τῇ τυράννῳ μακαρίᾳ νύμφῃ δότε | φέροντες· οὔτοι δῶρα μεμπτὰ δέξεται). As has been observed by Cazzaniga, the Ciris context finds a surprisingly close parallel in Ariadne’s speech in Nonnus (D. 47.392‒ 393 ὀλβίστῃ σέο νύμφῃ | τλήσομαι ὡς θεράπαινα πολύκροτον ἱστὸν ὑφαίνειν),⁵⁵ which would seem to point at a common source other than Euripides, though it is not impossible that Nonnus was familiar with Latin poetry.⁵⁶ In any event, such a promise of loyalty towards a potential future rival cannot but raise the suspicion that, like Medea, Scylla is not fully sincere. If this allusion is intended,

 Add also perhaps ‒ non equidem ex isto speraui corpore posse | tale malum nasci and Med. ‒ ἀνδρῶν δ’ ὅτῳ χρὴ τὸν κακὸν διειδέναι | οὐδεὶς χαρακτὴρ ἐμπέφυκε σώματι.  Cazzaniga () , cf. Lyne () .  See Trimble ()  with references.

5.7

163

a third possible explanation arises for Scylla’s being reasonably ‘polite’ to Minos: perhaps we can extrapolate the Euripidean paradigm to her entire monologue, which as a whole would parallel not so much Medea’s first, angry, speech to Jason, as her second one (869‒905), affectedly submissive (776 μαλθακοὺς λέξω λόγους).⁵⁷

5.7 As has been pointed out in an earlier section, Europa’s speech as she sails sitting on the bull’s back parallels the situation of Scylla’s lament (to quote again: 389 suspensa nouo ritu de nauibus, 148 ξείνην ναυτιλίην ἐφέπω) closer than any other such monologue of a ‘betrayed heroine’ does. And although Europa’s speech itself does not feature further specific points of contact with Scylla’s (it is much shorter, and after all a happy ending awaits Europa), the latter has a telling allusion to another context in Moschus’ poem (428‒432): tene ego plus patrio dilexi perdita regno? tene ego? nec mirum: uultu decepta puella (ut uidi, ut perii, ut me malus abstulit error!) non equidem ex isto speraui corpore posse tale malum nasci; forma uel sidera fallas.

I would argue that the motif of love at first sight can here be connected specifically, even if not exclusively, with Moschus’ treatment of the same trope (74‒79): ἦ γὰρ δὴ Κρονίδης ὥς μιν φράσαθ’ ὣς ἐόλητο θυμὸν ἀνωίστοισιν ὑποδμηθεὶς βελέεσσι Κύπριδος, ἣ μούνη δύναται καὶ Ζῆνα δαμάσσαι. δὴ γὰρ ἀλευόμενός τε χόλον ζηλήμονος Ἥρης παρθενικῆς τ’ ἐθέλων ἀταλὸν νόον ἐξαπατῆσαι κρύψε θεὸν καὶ τρέψε δέμας καὶ γείνετο ταῦρος.

To begin with, let us note that both contexts also share the motif of the deceitfulness of outward appearance: just as Zeus uses his disguise to entice Europa (παρθενικῆς τ’ ἐθέλων ἀταλὸν νόον ἐξαπατῆσαι), so Minos is able to seduce Scylla with his good looks (uultu decepta puella). Most importantly, however, ut uidi, ut perii can be argued to translate ὥς μιν φράσαθ’ ὣς ἐόλητο. Of course, line 430

 Cf. Schiesaro’s () ‒ argument for the importance of this scheming Medea as a model for Dido in Virgil.

164

5 An epic voyage (lines 386‒477)

is exactly identical with Verg. Ecl. 8.41, which in turn is no doubt related to Theocr. 2.82 χὠς ἴδον, ὣς ἐμάνην, ὥς μοι πυρὶ θυμὸς ἰάφθη and 3.42 ὡς ἴδεν, ὣς ἐμάνη, ὣς ἐς βαθὺν ἅλατ’ ἔρωτα. No wonder that Virgil’s indisputable connection to the Theocritean context(s) has been used as an argument for the priority of the Eclogues over the Ciris. ⁵⁸ But perii, I would contend, is a translation of Moschus’ ἐόλητο rather than of Theocritus’ ἐμάνη(ν). As has already been observed by Skutsch,⁵⁹ this use of perii finds a striking parallel in Achilles Tatius (1.4.4): ὡς δὲ εἶδον, εὐθὺς ἀπωλώλειν; what Skutsch failed to notice is that Achilles Tatius alludes here to none other than Moschus⁶⁰ and that ἀπωλώλειν is evidently an interpretation of ἐόλητο. The form ἐόλητο is extremely rare (before Moschus it only occurs in Apollonius at 3.471, itself probably based on Pind. Pyth. 4.233 ἐόλει) and of uncertain etymology.⁶¹ Although the scholia and lexica normally gloss ἐόλητο with ἐτετάρακτο or similar,⁶² it is far from unlikely that it could also have been understood by the ancient readers, because of its sound if for no other reason, as a form of ὄλλυσθαι. Both Achilles Tatius on the one hand and the Ciris (and Virgil) on the other would seem to follow precisely this interpretative tradition. If, as I am arguing, ut uidi, ut perii is indeed an allusion to Moschus’ ὥς μιν φράσαθ’ ὣς ἐόλητο, can this be considered proof that the Ciris predates Virgil? If we followed the logic of most earlier attempts at solving the Cirisfrage, it would certainly be a very strong argument for the Ciris’ priority. But, as I have already tried to show on several occasions, the underlying assumption that normally a poet like Virgil or the author of the Ciris would only refer to one model at a time does not merely underestimate the sophistication of this sort of poetry, but misses a fundamental point of its poetics. On the contrary, the normal practice would be for a poet to take into account as many instances of a given topos as possible. Accordingly, even if it may be convenient to say that Moschus is here the primary model, the tripartite structure of the line certainly reflects the analogous structure found in the two Theocritean contexts. Moreover, the nominative case of error, a word thought to be a pun on ἔρωτα in Theocritus,⁶³ may actually be intended to evoke the nominative ἔρως in Il. 14.294 ὡς δ’ ἴδεν, ὥς μιν ἔρως πυκινὰς φρένας ἀμφεκάλυψεν, the ultimate Homeric prototype of all the similar

 See especially Jahn () ‒. Gall () ‒ objects that the Ciris is no less likely to allude to Theocritus.  Skutsch ()  n. .  On Achilles Tatius’ use of Moschus, see Mignogna () ‒, with further references.  For a discussion, see Braswell () ‒; cf. Campbell () .  For a collection of examples, see Bühler () ‒.  Cf. Wills ()  n. .

5.7

165

passages. Whether ut uidi, ut perii, ut me malus abstulit error was originally written for the Ciris or the Eclogues, we can reasonably assume that the poet who wrote it (as well as the one who borrowed it) was thinking of at least these four Greek poetic contexts if not more (two in Theocritus, one in Moschus and one in Homer). It is worth noting that precisely this Homeric context deals in an emphatic manner with the issue of typicality, as Zeus describes his present feelings for Hera by comparing them with what he has felt for many other goddesses and women (14.315‒328). It is primarily for that reason, and perhaps not so much because Europa is mentioned in the ‘catalogue’ of Zeus’ love affairs (14.321‒322 οὐδ’ ὅτε Φοίνικος κούρης τηλεκλειτοῖο, | ἣ τέκε μοι Μίνων τε καὶ ἀντίθεον Ῥαδάμανθυν), that Moschus alludes to Homer.⁶⁴ This reading of Moschus’ compositional strategy is further substantiated by an allusion to Callimachus, as the line ending ζηλήμονος Ἥρης ‘quotes’ Zeus speaking of his readiness to commit adultery with other goddesses (H. 3.29‒31): ὅτε μοι τοιαῦτα θέαιναι | τίκτοιεν, τυτθόν κεν ἐγὼ ζηλήμονος Ἥρης | χωομένης ἀλέγοιμι. But, of course, Moschus’ interest in the issues of typicality and exemplarity is most obviously manifested on the internal level, in the Io ecphrasis (43‒61): the story of Io creates a perspective against which the main narrative has to be read.⁶⁵ When reading ὥς μιν φράσαθ’ ὣς ἐόλητο, or ὡς ἴδεν, ὣς ἐμάνη, or ut uidi, ut perii, we are expected, I suggest, to recall our previous reading experience in the same way as Zeus recalls his previous erotic experience at the moment he sees Hera. As I was proposing in an earlier chapter (3§7), love as a poetological metaphor can be understood in philosophical terms as a cognitive activity or capacity directed at grasping the typical and universal through the particular and singular, which constitutes a fundamental prerequisite for composing (but also reading) poetry. In the Homeric episode we observe a very similar process, as Zeus’ momentary passion for Hera makes him recall other analogous situations and brings him, as it were, to abstracting the notion of erotic passion as such (14.315 θεᾶς ἔρος [ἠ]δὲ γυναικός). In the Ciris – I suggest, this is a point we are supposed to infer – it is likewise an experience of love that makes Scylla speak in a language so deeply rooted in the poetic tradition. Before proceeding further, I would offer an additional indirect argument for accepting the allusion to Moschus in the Ciris. Although this is not generally acknowledged, the Scylla episode in Ovid’s Metamorphoses is full of references to  Note that ἐόλητο | θυμὸν ἀνωίστοισιν ὑποδμηθεὶς βελέεσσι points specifically at Il. .‒  οὐ γάρ πώ ποτέ μ’ ὧδε θεᾶς ἔρος οὐδὲ γυναικὸς | θυμὸν ἐνὶ στήθεσσι περιπροχυθεὶς ἐδάμασσεν, lines introducing the ‘catalogue’.  For a recent discussion, see Smart () ‒, with further references.

166

5 An epic voyage (lines 386‒477)

the Europa. First of all, a point also relevant to the Ciris, Minos is the offspring of precisely that union of Zeus and Europa (and of course, both Zeus and Minos sail to Crete, even if from opposite directions). Ovid alludes to this fact by referring to Minos as ducis Europaei (8.23), a spondaic verse-ending hinting at Moschus’ frequent positioning of Εὐρωπείη at the end of a line (41, 62, 93, 129).⁶⁶ Moreover, the Ovidian Scylla, in contrast to the heroine of the Ciris, twice mentions Minos’ parentage explicitly (8.49‒50 si quae te peperit talis, pulcherrime rerum, | qualis es, ipsa fuit, merito deus arsit in illa, 120 non genetrix Europa tibi est). For the sake of brevity, further parallels, some of which may be less convincing than others, will only be listed with minimal comments. Scylla watches the fighting from the walls (8.20 spectare ex illa rigidi certamina Martis), Europa sees the two continents fight for her in a dream (8 ὠίσατ’ ἠπείρους δοιὰς περὶ εἷο μάχεσθαι). Ovid and Moschus describe the onset of a night in similar terms (8.82‒84 nox interuenit tenebrisque audacia creuit. | prima quies aderat, qua curis fessa diurnis | pectora somnus habet, 2‒4 νυκτὸς ὅτε τρίτατον λάχος ἵσταται ἐγγύθι δ’ ἠώς, | ὕπνος ὅτε γλυκίων μέλιτος βλεφάροισιν ἐφίζων | λυσιμελὴς πεδάᾳ μαλακῷ κατὰ φάεα δεσμῷ). Minos wishes that earth and sea be refused to Scylla (8.98 tellusque tibi pontusque negetur!); by contrast, Zeus is admired by Europa for being able to tread both earth and sea (142‒143 σὺ δὲ χθόνα καὶ κατὰ πόντον | ἄτρομος ἀίσσεις). Both texts refer to Crete as Zeus’ birthplace, which will only receive Europa (158‒159 Κρήτη δέ σε δέξεται ἤδη | ἥ μ’ ἔθρεψε καὶ αὐτόν), but not Scylla (8.99‒100 certe ego non patiar Iouis incunabula, Creten, | qui meus est orbis, tantum contingere monstrum), though this is precisely her wish (8.118 nobis ut Crete sola pateret). Scylla finds herself alone as she comes to the shore (8.104‒105 Scylla freto postquam deductas nare carinas | nec praestare ducem sceleris sibi praemia uidit), Europa, as she sails away (131‒132 ἣ δ’ ὅτε δὴ γαίης ἄπο πατρίδος ἦεν ἄνευθεν, | φαίνετο δ’ οὔτ’ ἀκτή τις ἁλίρροθος οὔτ’ ὄρος αἰπύ). Scylla asks Minos, and Europa Zeus, about the direction in which they are sailing (8.108 quo fugis? 135 πῇ με φέρεις;). Scylla complains that Minos was not moved by her love (8.111‒112 nec te | noster amor mouit), Zeus explains that it was his love for Europa which prompted him to cross the sea (157 σὸς δὲ πόθος μ’ ἀνέηκε). Scylla questions the official version of Minos’ descent (8.122‒123 nec Ioue tu natus, nec mater imagine tauri | ducta tua est, contrast 155‒ 156 αὐτός τοι Ζεύς εἰμι, κεἰ ἐγγύθεν εἴδομαι εἶναι | ταῦρος, 158 ταύρῳ ἐειδόμενον). But despite all these pointed differences, Scylla is able in the end to join Minos’ sea voyage (8.138 me miseram! 141‒142 insequar inuitum puppemque amplexa recuruam | per freta longa trahar), so as to imitate Europa’s journey on the

 On spondaic verses as ‘figure of allusion’, see Wills () .

5.8

167

back of the bull (146‒148 ὤμοι ἐγὼ μέγα δή τι δυσάμμορος, ἥ ῥά τε δῶμα | πατρὸς ἀποπρολιποῦσα καὶ ἑσπομένη βοῒ τῷδε | ξείνην ναυτιλίην ἐφέπω καὶ πλάζομαι οἴη, 152 διέρχομαι ὑγρὰ κέλευθα).

5.8 Looking at Scylla’s monologue once again from a wider perspective, we may note that perhaps its most unusual part are the lines in which Scylla analyses her love for Minos (428‒437). As we have seen earlier, the preceding section (414‒427), which can be said to deal with the ‘objective’ past, is basically modelled on the two speeches of Medea in Apollonius. Likewise, the rest of Scylla’s monologue (438‒458), where she expresses her fears for the future and complains about her present situation, finds pointed parallels in the speech of the Catullan Ariadne. But the central ten lines (preceded by twenty-four and followed by twenty-one) are, if not entirely without points of contact with the speeches of other ‘betrayed heroines’, yet certainly very conspicuous in their detailed treatment of love, especially when viewed against their subtexts (one such subtext, in Moschus’ Europa, we just have discussed). It is normally the narrator who comments on the nature of love (as Apollonius does, 4.445 σχέτλι’ Ἔρως, μέγα πῆμα, μέγα στύγος ἀνθρώποισιν, and then Catullus, 64.94‒95). In the Ciris, however, the subject of love has already come out in the conversation of Scylla with her nurse, where Carme could claim a certain expertise in the matter (242‒243), whilst Scylla would confuse familial love with erotic love (259‒264). When discussing those earlier contexts, I suggested that they might contain certain philosophical or poetological connotations; now we can see that Scylla, too, has come, as it were, to a deeper understanding of love. Let us consider another subtext underlying the central part of Scylla’s monologue. Lines 431‒432 non equidem ex isto speraui corpore posse | tale malum nasci; forma uel sidera fallas, I would argue, allude to Bion fr. 14 (Reed): ἅμερε Κυπρογένεια, Διὸς τέκος ἠὲ θαλάσσας, τίπτε τόσον θνατοῖσι καὶ ἀθανάτοισι χαλέπτεις; τυτθὸν ἔφαν· τί νυ τόσσον ἀπήχθεο καὶ τὶν αὐτᾷ ταλίκον ὡς πάντεσσι κακὸν τὸν Ἔρωτα τεκέσθαι, ἄγριον, ἄστοργον, μορφᾷ νόον οὐδὲν ὁμοῖον; ἐς τί δέ νιν πτανὸν καὶ ἑκαβόλον ὤπασας ἦμεν ὡς μὴ πικρὸν ἐόντα δυναίμεθα τῆνον ἀλύξαι;

To start with, the idiomatic τυτθὸν ἔφαν, ‘moreover’, is remarkably close to – only a few lines below – 441 parua queror (confirming Housman’s emendation

168

5 An epic voyage (lines 386‒477)

of the transmitted magna). It can also be pointed out that the two contexts share the idea that ‘appearances deceive’ (forma uel sidera fallas, μορφᾷ νόον οὐδὲν ὁμοῖον: note that forma occupies the same metrical position as μορφᾷ, in contrast to the Catullan phrasing of this concept – 64.175‒176 celans dulci crudelia forma | consilia – which, however, is closer to Bion in the explicit opposition of ‘thoughts’ and ‘looks’). Furthermore, both contexts display the notion of the invincibility of love (ὡς μὴ πικρὸν ἐόντα δυναίμεθα τῆνον ἀλύξαι, 437 omnia uicit amor: quid enim non uinceret ille?). Yet the most striking parallel is that between tale malum nasci and ταλίκον ὡς πάντεσσι κακὸν τὸν Ἔρωτα τεκέσθαι. At first sight, this allusion may seem rather confusing. Bion speaks of Aphrodite giving birth to Eros, ‘a great evil to all’; the Ciris refers to a great evil born of Minos (or, to be precise, ex isto … corpore), but what great evil? The first impression is, and conceivably the Ciris poet may have expected the readers to have it, that Scylla means Minos’ cruelty to herself. Admittedly, it is not too forced a figure to speak of one’s actions as, literally, born from one’s body, but in the present context such a reading raises suspicion. It seems difficult, immediately after 429 uultu decepta puella, to take isto corpore to refer abstractly to the person of Minos rather than his outward appearance. What is produced by Minos’ beauty is nothing other than malus error, which, I believe, is the actual referent of tale malum. As pointed out above, error can plausibly be taken here as a ‘translation with paronomasia’⁶⁷ for ἔρως. Thus, in both Bion and the Ciris the ‘great evil’ will be love: in the former as a deity, in the latter as a psychological state. Both the speaker of Bion’s fragment and Scylla share the surprise that such an evil thing as love can be the offspring of what seems so different. Although the notion that love is caused by the sight of beauty may be a commonplace, and although Scylla could actually have no contact with Minos but the visual (175 speculator amorem), I suggest that such a pointed contrast between the (more or less) rationalist view of love that we find in the Ciris passage and the mythological one given in Bion’s fragment is intended to convey deeper implications. It is hardly irrelevant that for Lucretius erotic desire is similarly stimulated by visual images, simulacra. As Brown points out, this Epicurean “theory may be viewed as a rationalization of the commonplace idea that desire and love are caused by the sight of beauty.”⁶⁸ We already have noted in the Ciris several allusions to Lucretius’ discussion of love, and although no textual parallel occurs in the present context, it is not too unlikely an assumption that Lucretius may be implied here as well. Brown also makes another relevant point, as he

 O’Hara’s ()  term.  Brown () .

5.9

169

proposes “to contrast Lucretius’ reduction of this notion to the physical action of images and semen with Plato’s idealistic belief that the sight of beauty stirs something deep within the soul.”⁶⁹ This structural analogy between Plato’s and Epicurus’ views of the mechanics of love, with their ethical attitudes to the phenomenon being radically opposed, is no isolated case. In fact, from a certain perspective, labelling Epicurus a simius Platonis would not be too much off mark.⁷⁰ As I already have argued on several occasions, the Ciris poet is particularly interested in highlighting and exploring this sort of contradiction between the world-views expounded in his different models, and I would suggest that something similar is going on here.

5.9 That the subject of love is treated in Scylla’s monologue within an implicitly philosophical perspective can further be confirmed by another striking allusion (433‒437): me non deliciis commouit regia diues, curalio fragili aut electro lacrimoso, me non florentes aequali corpore nymphae, non metus impendens potuit retinere deorum: omnia uicit amor: quid enim non uinceret ille?

Of course, the last line of the quoted passage is well known to have a parallel in, and is often assumed to be modelled on, Virgil’s tenth eclogue (10.69 omnia uincit Amor: et nos cedamus Amori). I would, however, argue that the Ciris actually alludes here to Callimachus’ hymn to Zeus (1.70‒75): εἵλεο δ’ αἰζηῶν ὅ τι φέρτατον· οὐ σύ γε νηῶν ἐμπεράμους, οὐκ ἄνδρα σακέσπαλον, οὐ μὲν ἀοιδόν· ἀλλὰ τὰ μὲν μακάρεσσιν ὀλίζοσιν αὖθι παρῆκας ἄλλα μέλειν ἑτέροισι, σὺ δ’ ἐξέλεο πτολιάρχους αὐτούς, ὧν ὑπὸ χεῖρα γεωμόρος, ὧν ἴδρις αἰχμῆς, ὧν ἐρέτης, ὧν πάντα· τί δ’ οὐ κρατέοντος ὑπ’ ἰσχύν;

 Brown ()  n. .  Cf. De Lacy () : “The world view that Lucretius so eloquently proclaims in De Rerum Natura is at almost every point antithetical to Platonism, and often the antithesis is so precise that passages in Nat. can be paired with passages from the Dialogues. […] [Lucretius] not only rejected Platonism but even derived anti-Platonic arguments from the Dialogues, thus turning Plato against himself.”

170

5 An epic voyage (lines 386‒477)

The parallelism of expression between the concluding lines of the two passages is evident, even if Callimachus’ formulation is somewhat more subtle. In the Ciris we find a fairly straightforward statement that ‘love conquered everything, for what could love not conquer?’. Callimachus starts similarly by asserting that ‘everything is under the kings’ power’, but then inconspicuously switches from plural to singular: ‘for what is not under the ruler’s power?’.⁷¹ As McLennan aptly comments, “κρατέοντος appears in this sedes at Theocritus, 22, 213, where it refers to Zeus, as it probably does here (in addition to the πτολίαρχοι).”⁷² Thus, on the one hand, Zeus, of whom Callimachus speaks as the supreme power, is in the Ciris paralleled by the all-conquering power of love, or Amor; but on the other, the point is even more subtle, for the allusion can be taken as a ‘correction’ of Callimachus’ claim, implying that love actually overpowers even Zeus (which is, of course, a common topos of love poetry). However, the similarities between the two passages are not confined to their concluding lines, as both are elaborate priamels. The Ciris offers, again, a more straightforward variant, though it is not as artless as it may seem at first sight, for the gradation – from wealth (regia diues) to friends (nymphae) to religion (metus deorum), all proving powerless against love, – is no doubt intended. Callimachus likewise names three professions not chosen by Zeus: the sailor, the soldier, and the poet. Here, I suggest, there is a gradation of honour, as trade is a less honourable occupation than military service, which in turn is less honourable than poetry. The same list, with the farmer substituted for the poet, is then repeated in reverse order, referring to those subject to the king’s power. Now the principle of gradation is apparently different: the farmer is the least mobile and therefore the least independent of the three, the sailor the most mobile and independent, the soldier in between, but all three are equally under the king’s control. It seems also worth noting that the second list is summed up by the abstract πάντα, answered by τί (instead of πάντες and τίς, which would perhaps be more natural): for these we find in the Ciris exact equivalents in omnia and quid. This shift from πάντες to πάντα (similarly to that from plural to singular in κρατέοντος) is no doubt intended to signal the universality of the hierarchical principle, which should

 On the formal level, it should be noted that this sort of rhetorical questions (quid enim non uinceret ille? τί δ’ οὐ κρατέοντος ὑπ’ ἰσχύν;) is characteristic of both the Ciris and Callimachus. In the Ciris it is often assumed to originate from Ovid, who likewise has an exceptional liking for the device (see Munari [] ‒, cf. Lyne [] ‒), but there is no reason why it should not stem directly from Callimachus.  McLennan () . Theocr. . αὐτοί τε κρατέουσι καὶ ἐκ κρατέοντος ἔφυσαν: a line that emphasises a certain equivalence of the earthly rulers (κρατέουσι) and the supreme divine ruler (κρατέοντος) by linguistic means.

5.9

171

thus apply not only to different social ranks, but to – everything. In other words, Callimachus seems to be asserting a philosophical principle rather than merely speaking in mythological terms, and I would argue that this has not been lost on the Ciris poet. As has been convincingly demonstrated by Cuypers, Callimachus’ hymn to Zeus is systematically engaged with a hymn to Eros by Antagoras of Rhodes.⁷³ Of this latter hymn only a short fragment survives, which has no obvious points of contact with the Ciris; however, the substitution of love (or Amor) for Zeus as the supreme power in the allusion discussed above may imply that the Ciris poet was aware of Callimachus’ model. As Cuypers argues, it is possible “to see the Hymn to Eros as an Academic reply to the Stoic hymns to Zeus [identified with logos] – more precisely as a direct reply to Aratus by Antagoras, […] who selected the god who in Plato’s Symposium and Phaedrus leads to true knowledge as the Academic principle worthy of a hymn”;⁷⁴ in his turn, “Callimachus invites a reading of his poem against the background of the major philosophical controversy of his age: that between the Stoics and the Academics.”⁷⁵ It seems natural to assume that, mutatis mutandis, the same can be said of the Ciris when it states: omnia uicit amor. Whether or not a similar affirmation of the omnipotence of love was explicitly given in Antagoras’ hymn, this idea may actually be implied in Callimachus by means of another allusion (1.76‒79): αὐτίκα χαλκῆας μὲν ὑδείομεν Ἡφαίστοιο, τευχηστὰς δ’ Ἄρηος, ἐπακτῆρας δὲ Χιτώνης ᾿Aρτέμιδος, Φοίβου δὲ λύρης εὖ εἰδότας οἴμους· “ἐκ δὲ Διὸς βασιλῆες.”

As Cuypers has also shown, there are strong reasons “to read the Hymn to Zeus against the backdrop of Plato’s Symposium.”⁷⁶ I would suggest that this passage, following in the hymn immediately after the one quoted before, may specifically allude to Agathon’s argument in the Symposium that the invention of all arts and trades by different gods was inspired ultimately by Eros (197a–b): τοξικήν γε μὴν καὶ ἰατρικὴν καὶ μαντικὴν ᾿Aπόλλων ἀνηῦρεν ἐπιθυμίας καὶ ἔρωτος ἡγεμονεύσαντος, ὥστε καὶ οὗτος Ἔρωτος ἂν εἴη μαθητής, καὶ Μοῦσαι μουσικῆς καὶ Ἥφαιστος χαλκείας καὶ ᾿Aθηνᾶ ἱστουργίας καὶ “Ζεὺς κυβερνᾶν θεῶν τε καὶ ἀνθρώπων.”

   

Cuypers Cuypers Cuypers Cuypers

(). () . () . () .

172

5 An epic voyage (lines 386‒477)

I would in particular emphasise that both lists end with a reference to Zeus and, moreover, that in both cases this reference is a quotation. Like Callimachus, Agathon attributes to Zeus the ‘art’ of ruling over gods and men, but, unlike Callimachus, he also asserts that Zeus learnt it from Eros.

6 The metamorphosis (lines 478‒541) 6.1 In this final episode the Ciris comes to its telos as it achieves its aetiological objective stated in the proem, the appearance of a new bird species (48‒51): impia prodigiis ut quondam exterrita amoris Scylla nouos auium sublimis in aere coetus uiderit et tenui conscendens aethera penna caeruleis sua tecta super uolitauerit alis.

Scylla’s metamorphosis is thus not a mere by-product of what seems the main story, but in fact very much its specific goal. Accordingly, it is narrated in great detail: first the events that directly lead to it (478‒489), then the actual process of transformation (490‒507), and finally its consequences (510‒541).¹ We should further note that the outcome of the story does not only exist within the mythological universe of the poem itself, but is (assumed to be) perpetuated within our present-day world (the pursuit of the ciris by the sea eagle). Such an emphasis on the final result counterbalances the poet’s interest in the issues of initiality and causality that I highlighted when discussing the first narrative part of the Ciris. Given the programmatic importance of Scylla’s transformation into the ciris, it seems indeed surprising to find “two clear and distinct conceptions of metamorphosis at play (rescue from suffering, punishment) which, one might have thought, were mutually exclusive”; although “to an extent the poet has reconciled them”, there remain inconsistencies in the plot which Lyne interprets as “a direct sign that he is contaminating different sources.”² It is, of course, difficult to deny that the Ciris poet was familiar with more than one source; but the admission into the poem of different, sometimes apparently incompatible, variants is, I shall be arguing, not only deliberate, but in fact fundamental to his poetological concerns.

 As Lyne ()  observes, “Ovid, unusually for him, skirts describing metamorphosis at Met.  (Scylla). ff. One assumes for a good reason: perhaps because it had been recently done.” The most natural assumption would be that Ovid is reacting to the Ciris.  Lyne () .

174

6 The metamorphosis (lines 478‒541)

6.2 In the preceding chapter I have extensively argued that Scylla is cast as a poet figure, and in particular that her forced sea voyage is meant to evoke the metaphor of the ship of song. Here we shall see further development of such poetological tropes and motifs, though the emerging picture will be more complex. We may begin with a rather unexpected parallel for the comparison of Scylla being dragged behind the stern of Minos’ ship with a dinghy following a larger ship in storm (478‒480): fertur et incertis iactatur ad omnia uentis, cumba uelut magnas sequitur cum paruula classis Afer et hiberno bacchatur in aequore turbo.

As has been pointed out by Hollis, this passage is likely to be modelled on a context in Nicander’s Theriaca where the same simile describes the crooked motion of the cerastes (266‒270):³ αὐτὰρ ὅ γε σκαιὸς μεσάτῳ ἐπαλίνδεται ὁλκῷ, οἶμον ὁδοιπλανέων σκολιὴν τετρηχότι νώτῳ, τράμπιδος ὁλκαίης ἀκάτῳ ἴσος ἥ τε δι’ ἅλμης πλευρὸν ὅλον βάπτουσα κακοσταθέοντος ἀήτεω, εἰς ἄνεμον βεβίηται ἀπόκρουστος λιβὸς οὔρῳ.

In turn, the Nicandrian passage has been argued to be a reversed imitation of a comparison in Apollonius assimilating the Argo’s chaotic progress through the Tritonian lake to the motion of a snake suffering from the sun (4.1537‒1541, 1546‒1547):⁴ ἀλλ’ ὅτε δή ῥ’ ἐπὶ νηὸς ἔβαν, πρήσσοντος ἀήτεω ἂμ πέλαγος νοτίοιο, πόρους τ’ ἀπετεκμαίροντο λίμνης ἐκπρομολεῖν Τριτωνίδος, οὔ τινα μῆτιν δὴν ἔχον, ἀφραδέως δὲ πανημέριοι φορέοντο. ὡς δὲ δράκων σκολιὴν εἱλιγμένος ἔρχεται οἶμον, … ὣς ᾿Aργὼ λίμνης στόμα ναύπορον ἐξερέουσα ἀμφεπόλει δηναιὸν ἐπὶ χρόνον.

 Hollis () ‒.  See Jacques () CXIII n. , Magnelli () ‒, Overduin () .

6.2

175

It is obviously tempting to see in the Ciris context a ‘window’ allusion through Nicander to Apollonius, even if the only textual parallel with the Argonautica is not very close (fertur et incertis iactatur ad omnia uentis, ἀφραδέως δὲ πανημέριοι φορέοντο). Indeed, this would be an expected continuation of the implicit analogy between Scylla and the Argo that we observed on several occasions in the preceding narrative section. The wider contexts do in fact provide some additional points of contact: both passages belong to the accounts of either Minos’ or Jason’s return sea journeys, there is an imminent danger of death for both Scylla and the Argonauts, and in both cases help comes from a sea deity, either Neptune’s wife Amphitrite or their son Triton. The Apollonian allusion is therefore easy to make sense of, even though its reality can be questioned; but how should we interpret the parallel with Nicander, which seems more certain, but also less readily comprehensible? The implicit comparison of Scylla with a snake, especially in a context that emphasises her beauty (481 tale decus formae), is definitely puzzling. A partial answer may be that the Ciris poet had a certain inclination towards the grotesque, and indeed there are other cases of allusion to similarly ironic subtexts which undermine the pathos of the surface narrative. For example, the portrayal of Nisus earlier in the poem has two such subtexts that at first sight appear subversive (120‒ 122, 126‒128): nam capite a summo regis, mirabile dictu, candida caesaries florebat tempore utroque, at roseus medio surgebat uertice crinis. … ergo omnis caro residebat cura capillo, aurea sollemni comptum quem fibula ritu Cecropiae tereti nectebat dente cicadae.

One is Catullus’ somewhat sarcastic depiction of the Parcae in their senile infirmity (64.307‒310: note corpus tremulum): his corpus tremulum complectens undique uestis candida purpurea talos incinxerat ora, at roseae niueo residebant uertice uittae, aeternumque manus carpebant rite laborem.

However, as we have already seen in an earlier chapter, this allusion may have a more serious implication as it draws a certain analogy between Nisus’ purple lock and the Parcae’s thread. Moreover, there may also be a literary-historical dimension to it, for the Ciris seems to be making simultaneously a ‘window’ allu-

176

6 The metamorphosis (lines 478‒541)

sion to a possible model of the Catullan passage, a fragment of an otherwise unknown poem by Asius of Samos describing the fashion of the ancient Samians (fr. 13.1‒6):⁵ οἳ δ’ αὔτως φοίτεσκον ὅπως πλοκάμους κτενίσαιντο εἰς Ἥρης τέμενος, πεπυκασμένοι εἵμασι καλοῖς, χιονέοισι χιτῶσι πέδον χθονὸς εὐρέος εἶχον· χαῖται δ’ ᾐωρεῦντ’ ἀνέμῳ χρυσέοις ἐνὶ δεσμοῖς, χρύσειαι δὲ κορύμβαι ἐπ’ αὐτῶν τέττιγες ὥς· δαιδάλεοι δὲ χλιδῶνες ἄρ’ ἀμφὶ βραχίοσ’ ἕσαντες.

As we can see, both the Catullan passage and Asius’ fragment feature more or less the same progression from garments to hairstyle to hands (arms). More specifically, both the Parcae and the Samians are said to be tightly wrapped (complectens undique uestis, πεπυκασμένοι εἵμασι καλοῖς) in their ankle-length white clothes (candida … talos incinxerat, χιονέοισι χιτῶσι πέδον χθονὸς εὐρέος εἶχον⁶), with some sort of bands on their heads (roseae … uittae, χρυσέοις ἐνὶ δεσμοῖς). All this is hardly enough to posit a connection between Catullus and Asius, especially as we have no idea of the fragment’s wider context. But with the Ciris it has a very distinctive parallel: both the Samians and Nisus use the same specific type of hair clasps or brooches, the so-called cicadas (τέττιγες).⁷ That both texts also make special mention of the men’s lush manes (χαῖται δ’ ᾐωρεῦντ’ ἀνέμῳ, caesaries florebat) may admittedly only add little weight to the parallel, as it is a characteristic trait of that hairstyle in general. Yet there is another intriguing point of contact: the Samians are said to have been dressed in this manner when they went to Hera’s sanctuary, and it was likewise in Juno’s sanctuary that Scylla committed her offence – a fact told straight after the presentation of Nisus. It is conceivable, though only a speculation, that Asius went on to describe a religious ceremony in Hera’s honour which might have had further parallels with the Ciris (we may note the contrast between the Samians wrapped stately in their clothes, πεπυκασμένοι εἵμασι καλοῖς, and Scylla loosening hers, 145 sinus aquilone relaxans, cf. ᾐωρεῦντ’ ἀνέμῳ). If this is indeed a ‘window’ allusion through Catullus to Asius, it has an obvious structural similarity to what seems to be an allusion through Nicander to Apollonius, with the in-between intertext being in both cases somewhat ironic. A possible answer could

 For a discussion of the fragment, which is certainly not without problems, both textual and interpretative, see O’Sullivan ().  The meaning (and text) of the line is disputed, but it probably refers to the notion of Ἰάονες ἑλκεχίτωνες (Il. .).  For a rich collection of references to the τέττιγες, see Cook () ‒ n. .

6.3

177

therefore be that the use of Nicander (Catullus) may in the first place have been intended to acknowledge his use of Apollonius (Asius), who would be the main target of the allusion. The second subtext is the detailed, almost anatomically precise, portrait of Zetes and Calais in the catalogue of the Argonauts in Apollonius (1.219‒223): τὼ μὲν ἐπὶ κροτάφοισι ποδῶν θ’ ἑκάτερθεν ἐρεμνὰς σεῖον ἀειρομένω πτέρυγας, μέγα θάμβος ἰδέσθαι, χρυσείαις φολίδεσσι διαυγέας· ἀμφὶ δὲ νώτοις κράατος ἐξ ὑπάτοιο καὶ αὐχένος ἔνθα καὶ ἔνθα κυάνεαι δονέοντο μετὰ πνοιῇσιν ἔθειραι.

On the formal level, we may point out the exact rendering of κράατος ἐξ ὑπάτοιο with capite a summo and that of the parenthetic μέγα θάμβος ἰδέσθαι with mirabile dictu (though note the variatio, as uisu would be equally possible). Turning now to the meaning of this allusion, we can describe it, on the one hand, as similarly ironic or subversive, for indeed Nisus’ grandeur is undermined by the implicit comparison with this description of the Boreads, an embodiment of levity in both the literal and figurative sense. But on the other hand, the Apollonian subtext may also have a more serious purpose, namely to prefigure Nisus’ metamorphosis into a bird (note σεῖον ἀειρομένω πτέρυγας) and perhaps in particular his pursuit of the ciris: the most notable appearance Zetes and Calais make in Apollonius is when they chase the Harpies away from Phineus (2.273‒283). Looking back at the allusion to Nicander’s Theriaca, I would suggest that its raison d’être may likewise be not merely to cast a shadow of irony over the figure of Scylla, but more specifically to evoke her monstrous namesake – for what purposes, we shall see in the next section.

6.3 In an earlier chapter (2§5) I have already noted a similar allusion which links the two Scyllas directly, even if implicitly (130‒131 Scylla nouo correpta furore, | Scylla… referring to Ap. Arg. 4.827‒828 ἠὲ παρὰ Σκύλλης στυγερὸν κευθμῶνα νέεσθαι, | Σκύλλης…). The passage describing the process of Amphitrite’s decision to transform Scylla into a bird rather than a fish, I would suggest, makes an implicit reference to the Homeric Scylla, too (484‒488): sed tamen aeternum squamis uestire puellam infidosque inter teneram committere pisces non statuit (nimium est auidum pecus Amphitrites):

178

6 The metamorphosis (lines 478‒541)

aeriis potius sublimem sustulit alis, esset ut in terris facti de nomine ciris.

The statement that pecus Amphitrites may be dangerous is somewhat puzzling: why should Amphitrite’s ‘sheep’ – one imagines, peacefully grazing on seaweed – be of any danger even to a small fish? As I argue in greater detail elsewhere,⁸ this context pointedly alludes to Circe’s description of Scylla in the Odyssey (12.95‒97): αὐτοῦ δ’ ἰχθυάᾳ, σκόπελον περιμαιμώωσα, δελφῖνάς τε κύνας τε καὶ εἴ ποθι μεῖζον ἕλῃσι κῆτος, ἃ μυρία βόσκει ἀγάστονος ᾿Aμφιτρίτη.

Once we take into account this Homeric passage, it becomes obvious that the apparently innocent ‘Amphitrite’s livestock’ are no other creatures than ‘dolphins, sea dogs, and even bigger monsters’, ἃ μυρία βόσκει ἀγάστονος ᾿Aμφιτρίτη (note that the spondaic verse-ending Amphitrites functions as a ‘figure of allusion’ evoking the Greek σπονδειάζων⁹). The specific point may be that these dangerous sea beasts, which fall victim to the Homeric Scylla, would be particularly eager to take revenge on her unprotected namesake. Alternatively, since certain rationalising interpretations take ‘dolphins, sea dogs, and even bigger monsters’ as part of the dangerous natural phenomenon underlying Homer’s account of Scylla, it turns out to be the Homeric monster herself that Amphitrite saves Nisus’ daughter from. This latter reading may in fact be preferable, for it adds an important poetological implication: Amphitrite’s rescuing one Scylla from the other can be said to mirror the author’s rejection of that version of the story in which the Megarian princess is transformed into the monster. As pointed out earlier (5§3), the passage narrating Amphitrite’s decision to turn Scylla into a bird rather than a fish features an internal reference to a context in the proem announcing the narrator’s choice of the particular version that he will follow (90‒91 potius liceat notescere cirin | atque unam ex multis Scyllam non esse puellis). Now we can take this observation a step further. As we learn from another context in the proem (70‒76), one of the explanations of the origin of the Homeric Scylla claims that she was likewise a girl, with whom Neptune had committed adultery and whom Amphitrite had punished by turning her into a hideous beast. So I would suggest that although on the explicit level only two alternative ways of

 Kayachev ().  On spondaic verses as allusion signposts, cf. Wills () .

6.3

179

metamorphosing Nisus’ daughter are considered by Amphitrite, the other girl’s fate is hinted at by the reference to the sea goddess as coniunx Neptunia (483): since the Megarian Scylla did not have a love affair with Neptune, she is transformed into a beautiful bird rather than a monster. The above considerations may also offer a clue on 508‒509 et tamen hoc demum miserae succurrere pacto | uix fuerat placida Neptuni coniuge dignum (note that Amphitrite is again referred to as ‘Neptune’s wife’), in which Lyne saw “the poet’s visible effort to reconcile and fit into one and the same poem two basically different versions of the metamorphosis: (1) that it was a rescue; (2) that it was a punishment.”¹⁰ The function of these lines seems in fact more subtle. Why is transforming Scylla into a bird deemed ‘unworthy’ of Amphitrite? Lyne supposes it is because “being both a goddess and on Scylla’s side she might have done something better.”¹¹ I would suggest the narrator is actually questioning the sincerity of Amphitrite’s motives: had she been truly benevolent (placida) to Scylla, she would have done something better, but as it is, what she did was, in effect, eliminate a potential rival (note that one of the consequences of Scylla’s metamorphosis is that she will never be married: 512 non thalamus Syrio fragrans accepit amomo). In other words, Nisus’ daughter actually shares, to a degree, the fate of the other Scylla. How does this development in understanding Amphitrite’s motives conform with my suggestion that her role mirrors that of the narrator? Perhaps a qualification can be made: Amphitrite may be not so much entirely unwilling to help as, rather, unable to master her jealousy. The mixed character of Amphitrite’s motivation is precisely what sheds light on the nature of poetic composition: whilst on the explicit level after the proem the narrator consistently avoids mentioning the other Scylla, on the level of allusion she is never completely forgotten. But why is the Ciris poet unwilling, or indeed unable, to stop thinking of the Homeric monster? The answer seems to be that in this way he acknowledges the fact that thematic associations, which play a fundamental role in writing poetry, cannot be brought under full control. In earlier chapters we have had several occasions to observe in detail how he employs the technique of ‘composition by theme’, and now we see him reveal its darker, destructive, side. The Ovidian Scylla provides a good illustration, almost a commentary (Met. 14.59‒63): Scylla uenit mediaque tenus descenderat aluo, cum sua foedari latrantibus inguina monstris aspicit; ac primo, credens non corporis illas

 Lyne () ‒.  Lyne () .

180

6 The metamorphosis (lines 478‒541)

esse sui partes, refugitque abigitque timetque ora proterua canum. sed quos fugit attrahit una.

As Hardie remarks, here “the issue of believability, credulitas is transferred from the poet’s readers to the subject of metamorphosis herself.”¹² I would slightly change the emphasis and suggest that Scylla’s initial reaction can be compared with the unwillingness of the reader, or indeed of the author himself, to accept unwelcome intertextual associations as part of the text (credens non corporis illas | esse sui partes): yet in whatever way you may flee them, there is no escape (sed quos fugit attrahit una). In later sections we shall return to this issue and consider in what terms it is conceptualised in the Ciris.

6.4 Let us turn now to the account of the actual process of Scylla’s metamorphosis, the thoroughness of which is hardly paralleled in ancient poetry. Despite this thoroughness, however, or perhaps rather because of it, an attentive reader may have difficulty in getting a clear picture of exactly what is taking place. On the literal level, of course, we witness Scylla’s body changing physically before our eyes as she is being towed behind Minos’ ship (493‒495): sic liquido Scyllae circumfusum aequore corpus semiferi incertis etiam nunc partibus artus undique mutabant atque undique mutabantur.

Yet this very passage is preceded by a striking simile that compares Scylla’s transformation to the development of the embryo within the egg (490‒492): hic uelut in niueo tenera est cum primitus ouo effigies animantis¹³ et internodia membris imperfecta nouo fluitant concreta calore.

In the first place, this simile invites looking at the metamorphosis as something natural, even if exceptional, rather than blatantly miraculous. Indeed, its technical language is strongly reminiscent, as has been amply documented by Pi-

 P. Hardie (b) .  Lyne’s text seems preferable: tenerae cum primitus ouo | effigies animantur (note Greek parallels for the possibly technical use of animare in the next footnote).

6.4

181

geaud, of the terminology used in the account of embryogenesis in the Hippocratic treatise De natura pueri. ¹⁴ On the other hand, it opens the door to the suspicion that, instead, a rationalising explanation of a kind may be implied and, therefore, the actual description should not be taken literally. Another crucial subtext is provided by one of Lucretius’ arguments against the immortality of the soul (3.526‒539): denique saepe hominem paulatim cernimus ire et membratim uitalem deperdere sensum; in pedibus primum digitos liuescere et unguis, inde pedes et crura mori, post inde per artus ire alios tractim gelidi uestigia leti. scinditur atque animae haec quoniam natura nec uno tempore sincera existit, mortalis habendast. quod si forte putas ipsam se posse per artus introsum trahere et partis conducere in unum atque ideo cunctis sensum diducere membris, at locus ille tamen, quo copia tanta animai cogitur, in sensu debet maiore uideri; qui quoniam nusquamst, ni mirum, ut diximus ante, dilaniata foras dispargitur, interit ergo.

The hypothetical suggestion that, as the limbs gradually lose the capacity for sensation, the soul should shrink and gather together in one place in the body is alluded to, with a comic effect it may seem, in the description of Scylla’s pretty face turning into a bird’s beak (496‒498): oris honos primum et multis optata labella et patulae frontis species concrescere in unum coepere et gracili mentum producere rostro.

But this apparently humorous allusion also implies a serious question: is it at all possible that Scylla’s soul should survive such a radical change that affects her body? Even more sombrely, the transformation of Scylla’s legs into those of a bird

 Pigeaud (). I would particularly note the following parallels: uelut in niueo … ouo … sic liquido Scyllae circumfusum aequore corpus cf.  οἷον εἴ τις ὠοῦ ὠμοῦ τὸ ἔξω λεπύριον περιέλοι, ἐν δὲ τῷ ἔνδον ὑμένι τὸ ἔνδον ὑγρὸν διαφαίνοιτο; tenerae … effigies [sc. membrorum] cf.  τὰ εἴδεα τῶν μελέων … ἐπισκληρότερα γίνεται; internodia … concreta calore cf.  τὰ ὀστέα σκληρύνεται ὑπὸ τῆς θέρμης πηγνύμενα; effigies animantur cf.  ἡ δὲ σὰρξ αὐξομένη ὑπὸ τοῦ πνεύματος and τουτέων [sc. ἐντέρων] δὲ διαρθροῦται ὑπὸ τῆς πνοῆς ἕκαστα· φυσώμενα γὰρ διίσταται ξύμπαντα κατὰ συγγένειαν.

182

6 The metamorphosis (lines 478‒541)

is cast against the background of Lucretius’ scrupulous depiction of the legs gradually dying off (505‒507): inde alias partes minioque infecta rubenti crura noua macies obduxit squalida pelli et pedibus teneris unguis affixit acutos.

The allusion is further confirmed, and developed, by an internal reference, in the lines just preceding the quoted passage (503‒504 [pluma] marmoreum uolucri uestiuit tegmine corpus | lentaque perpetuas fuderunt bracchia pennas), to an earlier context in which Scylla voices her presentiment of an imminent death (449‒ 450): et caput inflexa lentum ceruice recumbit, | marmorea adductis liuescunt bracchia nodis. What is especially noteworthy is the use of liuescere, a rare verb that unmistakably points at precisely that context in Lucretius. In other words, the Ciris poet evidently wants his readers to view Scylla’s metamorphosis as something analogous to death, at least on a certain level. Now, if we link this with the implication of the egg simile, the result will be that the process of metamorphosis can be analysed into two stages: the dying of one living being and the coming into existence of another. This is also, we should note, exactly what happens to Nisus: first he dies and only later, after an interval, is he revived in the shape of a sea eagle (527 reddidit optatam mutato corpore uitam).¹⁵ What is therefore at issue, I suggest, is the actual mechanics of how Scylla can be thought to preserve her identity as she is transformed into the ciris. Since the continuity cannot be guaranteed by the body (whether or not it actually ‘dies’, it certainly becomes another body), it is the soul that takes over this task. In other words, from a philosophical perspective Scylla’s metamorphosis can be viewed as a case of metempsychosis, and indeed, as we shall see later on, this idea finds support in a number of subtexts. At the same time, as we just have seen, the Ciris alludes to a passage in Lucretius that argues specifically against the immortality of the soul. There are, in fact, at least two further analogous Lucretian contexts alluded to in the Ciris: one will be considered later, the other is Lucretius’ rather ironic discussion (3.726 in discrimen agendum) of whether it may be possible that pre-existing souls either seek bodies to live in, or build them themselves. Despite the difference of meaning in the two uses of discrimen, 499 qua se medium capitis discrimen agebat is likely to be pointing at that particular Lucretian context, the collocation discrimen ag- being extreme-

 Note the pointed ambiguity of mutato corpore: it can potentially mean both ‘with his body transformed’ (as in Verg. Georg. .) and ‘with his body replaced with another’ (as in Lucr. .).

6.5

183

ly rare in Latin verse (elsewhere only attested in Juvenal 6.500). The reference, again, may conceivably be to Lucretius’ argument against the immortality of the soul in general, but it is tempting to see the immediate context as a specific target. Here Lucretius is considering the spontaneous generation of maggots in a dead body: it is precisely the fact that the maggots’ souls are composed from particles of the soul previously belonging to the dead body that, somewhat paradoxically, proves its mortality. It is true that Scylla’s soul, as it passes into the ciris, would not have to suffer fragmentation in exactly the same way, but Lucretius’ earlier argument would certainly apply: the soul has to dissolve, in order to penetrate a new body (3.701 quod permanat enim, dissoluitur, interit ergo).

6.5 Let us nevertheless have a closer look at the subtexts that do suggest viewing Scylla’s metamorphosis as a form of metempsychosis, and consider what possibilities they raise for the interpretation of the Ciris. Perhaps the most expected subtext of this kind is Ennius’ account of his previous incarnations, even if individual textual parallels with the surviving fragments may not be conclusive. In general, Ennius’ is likewise a case, at one point, of human-to-bird transmigration (fr. 1.9 memini me fiere pauom). Furthermore, in Ennius the egg features prominently as an illustration of the doctrine of metempsychosis (fr. 1.8): oua parire solet genus pennis condecoratum, non animam. post inde uenit diuinitus pullis ipsa anima.

The egg simile of the Ciris seems evocative of this passage (note 491 animantis). The presence of the Ennian peacock in the allusive background of the Ciris can further be supported by a possible reference to another peacock, the one into which Argus is said to be transformed in Moschus’ Europa (58‒61): τοῖο δὲ φοινήεντος ἀφ’ αἵματος ἐξανέτελλεν ὄρνις ἀγαλλόμενος πτερύγων πολυανθέι χροιῇ, τὰς ὅ γ’ ἀναπλώσας ὡσεί τέ τις ὠκύαλος νηῦς χρυσείου ταλάροιο περίσκεπε χείλεα ταρσοῖς.

Since there are other allusions to Moschus (cf. 1§9, 3§4, 5§7), it is in itself plausible that this passage should provide a model for the description of Scylla’s metamorphosis. Some individual points of contact seem worth noticing, even if they may admittedly be too general to allow a firm conclusion (500‒504):

184

6 The metamorphosis (lines 478‒541)

ecce repente uelut patrios imitatus honores puniceam concussit apex in uertice cristam; at mollis uarios intexens pluma colores marmoreum uolucri uestiuit tegmine corpus lentaque perpetuas fuderunt bracchia pennas.

To start with a formal observation, in the Ciris we find three ‘golden lines’ in succession (502‒504); two can specifically be linked with similarly built five-word hexameters in the Europa: 503 (note uestiuit) with 61 (cf. περίσκεπε, similarly positioned) and 502 (uarios intexens pluma colores) with 59 (πτερύγων πολυανθέι χροιῇ). The latter parallel, commonplace as the notion of motley plumage may be, has more weight than is apparent at first sight, for in other contexts the ciris appears as a white bird (205 candida) with dark wings (51 caeruleis … alis): this is, I would suggest, a learned inconsistency,¹⁶ meant to signpost the allusion to Moschus. If, then, Scylla’s transformation into the ciris is in some way correlated to Homer’s reincarnation as a peacock, it is difficult to avoid the suspicion that it likewise bears a certain poetological implication, especially since Scylla herself is presented in a number of ways as a poet figure. Later on we shall consider what it means for the poet to have an immortal as opposed to a mortal soul, whilst for the moment I would address the issue of metamorphosis (metempsychosis) as a poetological trope from a slightly different perspective. As is noted, for example, by Hardie, in the Roman poetic tradition metamorphosis comes to be closely associated with metaphor and other tropes, a tendency culminating in “Ovid’s habit of generating tales of transformation out of a literalization of the figurative.”¹⁷ Indeed, metaphor and metamorphosis are similar in that they both posit an inherent identity between two seemingly disparate phenomena. If, for instance, a warrior is metamorphosed into a lion, we can say that what establishes the identity of the two creatures is the soul which passes from one body into the other; but what is the principle of identity behind the metaphor of a warrior as a lion? Generally speaking, it must be some abstract quality, or universal, shared by the two, for example courage or rage. Allusion, or indeed any kind of intertextual link between two contexts, likewise requires a tertium comparationis shared by the alluding passage and the passage alluded to. In the Ciris, I suggest, it is primarily allusion rather than metaphor (or any other figure of speech) that metamorphosis can be said to stand for. As we have already observed on several occasions, a particular subtext may, as it

 On deliberate inconsistencies as a literary device, see in general O’Hara ().  P. Hardie () .

6.5

185

were, prefigure either the actual course of Scylla’s (or Nisus’) transformation, or an alternative one which on the surface level the poet appears to reject, but which nevertheless is implicitly ‘narrated’ by means of allusion. What happens to the text as the reader recognises a subtext is indeed akin to metamorphosis; as we read of the Megarian Scylla and spot an allusion to her famous namesake, in our imagination she is in a sense identified with, even transformed into, the Homeric monster. Much as, in order for metamorphosis to take place, the soul must be able to transcend its body (that is, immortal), an image, to allow of a figurative use (or of an intertextual connection), requires a certain symbolic value that exceeds its literal meaning. In that sense, poetic (metaphoric) language can be said to be related to technical (literal) language in the same way as a Pythagorean universe (in which the soul outlives its body) is related to an Epicurean one (in which the soul dies with its body). This inherent quality of poetic discourse poses a paradox to an Epicurean attempting to write poetry, and it is precisely this paradox that the Ciris poet seems to be at pains to evince, if not solve. A further complication is that, once the symbolic potential of poetic language is recognised, there comes a danger of, so to speak, panpsychism, as ultimately anything can be symbolically linked with anything. To avoid chaos, the author (and later the reader) has to decide which metaphors or intertexts to give life to, much in the same way as he decides whether Scylla’s soul is to pass into a bird or a sea monster. The difficulty, however, is that the authorial decision on which intertexts to accept can never be final, for the last word is always the reader’s; it is to acknowledge this fact, I suggest, that the Ciris poet frequently introduces subversive allusions. Before moving on, I shall discuss one further subtext of this kind underlying the description of Scylla’s metamorphosis. It is Apollonius’ account of the Argonauts’ preparation for the attack of the Stymphalian birds and the attack itself (2.1069‒1072, 1083, 1088‒1089): ἀμφὶ δὲ χαλκείας κόρυθας κεφαλῇσιν ἔθεντο δεινὸν λαμπομένας, ἐπὶ δὲ λόφοι ἐσσείοντο φοινίκεοι. καὶ τοὶ μὲν ἀμοιβήδην ἐλάασκον· τοὶ δ’ αὖτ’ ἐγχείῃσι καὶ ἀσπίσι νῆα κάλυψαν. … ὡς δ’ ὁπότε Κρονίδης πυκινὴν ἐφέηκε χάλαζαν … ὣς πυκινὰ πτερὰ τοῖσιν ἐφίεσαν ἀίσσοντες ὕψι μάλ’ ἂμ πέλαγος περάτης εἰς οὔρεα γαίης.

The Apollonian context features at least two specific textual points of contact with the Ciris. First, since one would normally expect Nisus’ lock, or its counter-

186

6 The metamorphosis (lines 478‒541)

part, to be literally ‘purple’, puniceam cristam is a ‘figure of allusion’ (puniceus instead of purpureus) unmistakably evoking Apollonius’ λόφοι φοινίκεοι¹⁸ (cf. further ἐσσείοντο with concussit and ἐγχείῃσι καὶ ἀσπίσι νῆα κάλυψαν with uestiuit tegmine corpus). Second, the phrase perpetuas pennas closely renders πυκινὰ πτερά, with fuderunt possibly hinting at the storm simile (πυκινὴν ἐφέηκε χάλαζαν).¹⁹ This allusion is remarkable for more than one reason. To begin with, it continues the implicit association of Scylla with the Argo. Moreover, what happens with the Argo is likewise a transformation of a kind, almost a metamorphosis, into a horrendous monster. At the same time, Scylla, or rather the ciris, is in a sense assimilated to the dangerous Stymphalian birds shooting their brazen feathers. Whatever the tertium comparationis is that links the Apollonian model with its imitation in the Ciris (what we may also call the ‘soul’ of the allusion), it is never a fixed entity (or perhaps more correctly, our notion of it can never be a fixed one). Two allusions in Virgil’s Aeneid to the same Apollonian context can be said to be its further ‘reincarnations’, which, however, also affect its earlier ‘incarnation’ in the Ciris (assuming it predates the Aeneid: the chronological question is of no great importance at present). One is in a portrait of Turnus in battle (9.731‒733): continuo noua lux oculis effulsit et arma horrendum sonuere, tremunt in uertice cristae sanguineae clipeoque micantia fulmina mittit.

The verse-ending in uertice cristae attests to a direct connection with the Ciris (501 in uertice cristam),²⁰ whereas other details point to Apollonius: the enjambe-

 A similar case is Ciris  roseus crinis, likewise signalling the allusion to Catullus . (cf. above §). The point is obviously not that puniceus (or roseus) may be less appropriate than purpureus to describe the colour of Nisus’ hair, for which we have no eyewitness report. The reason why it is regularly described as purpureus is because this is the word used by Callimachus, Hec. fr. . (Hollis) πορφυρέην ἤμησε κρέκα, cf. Hollis () : “Although this fragment looks like a passing allusion, it may have sufficed to establish the purple lock as a famous poetic theme […]; the Latin poets regularly make it purple […] as does Nonnus.” In a comparable manner, Virgil’s mare purpureum (Georg. .), as well as Valerius’ sale purpureo (.), and Propertius’ purpureus arcus (..) are meant in the first place to evoke ἅλα πορφυρέην (Il. .) and πορφυρέην ἶριν (Il. .): cf. Edgeworth () on these and similar cases.  Note Lyne () : “[I]f we think of the poetical use of penna of an arrow […] and the quite common use of fundo of profuse shooting […] we have a context, a military one, in which fuderunt pennas could have a more natural function than in ours.”  Though, of course, we should bear in mind the possibility of a common source in, for example, Varro Atacinus’ Argonauts.

6.6

187

ment cristae | sanguineae reproduces that of λόφοι … | φοινίκεοι (note that sanguineae interprets φοινίκεοι as φοινοί), the initial position of horrendum corresponds to that of δεινόν, and the metaphor fulmina mittit may be a compression of the Apollonian simile (ἐφέηκε χάλαζαν – πτερὰ … ἐφίεσαν). The other is in the account of the attack of the two serpents on Laocoon (2.205‒208): incumbunt pelago pariterque ad litora tendunt; pectora quorum inter fluctus arrecta iubaeque sanguineae superant undas, pars cetera pontum pone legit sinuatque immensa uolumine terga.

Once more, we find iubaeque | sanguineae in enjambement. Like the Argo’s crew, some of whom were engaged in rowing and the rest in protecting the ship (καὶ τοὶ μὲν ἀμοιβήδην ἐλάασκον· | τοὶ δ’ αὖτ’ ἐγχείῃσι καὶ ἀσπίσι νῆα κάλυψαν), the serpents display one part of their bodies above the waves and with the other effect their progress (note pars cetera). Like the Stymphalian birds (ἀίσσοντες | ὕψι μάλ’ ἂμ πέλαγος περάτης εἰς οὔρεα γαίης), they head in the direction of the shore (incumbunt pelago pariterque ad litora tendunt). Much as our appreciation of the Ciris context is inevitably transformed as we learn that puniceam cristam is an imitation of Apollonius’ λόφοι φοινίκεοι, so it experiences further metamorphoses as we discover reuses of the same motif in Virgil’s cristae sanguineae and iubae sanguineae, with their even more sinister connotations. This cycle of metempsychosis can never be brought to an end.

6.6 Let us now consider what metempsychosis as a poetological trope means as applied to the figure of the poet. The obvious point to start from is Ennius’ claim to be a reincarnation of Homer (via a peacock). It is generally interpreted as merely a figurative way of saying that he follows, or indeed inherits, the Homeric epic tradition.²¹ As far as Ennius is concerned, this interpretation may be correct, but the idea of the poet as a sort of Pythagorean sage is likely to be Hellenistic in origin, and originally it would have had deeper implications. Although Apollonius never makes such a statement explicitly, it seems possible to argue that this concept is present, and plays an important role, in his Argonautica. I would suggest that the somewhat enigmatic figure of the herald Aethalides represents precisely that aspect of Apollonius’ poetics (1.640‒649):  See e. g. Aicher ().

188

6 The metamorphosis (lines 478‒541)

τείως δ’ αὖτ’ ἐκ νηὸς ἀριστῆες προέηκαν Αἰθαλίδην κήρυκα θοόν, τῷ πέρ τε μέλεσθαι ἀγγελίας καὶ σκῆπτρον ἐπέτραπον Ἑρμείαο σφωιτέροιο τοκῆος, ὅς οἱ μνῆστιν πόρε πάντων ἄφθιτον. οὐδ’ ἔτι νῦν περ ἀποιχομένου ᾿Aχέροντος δίνας ἀπροφάτους ψυχὴν ἐπιδέδρομε λήθη· ἀλλ’ ἥ γ’ ἔμπεδον αἰὲν ἀμειβομένη μεμόρηται, ἄλλοθ’ ὑποχθονίοις ἐναρίθμιος, ἄλλοτ’ ἐς αὐγὰς ἠελίου ζωοῖσι μετ’ ἀνδράσιν. ἀλλὰ τί μύθους Αἰθαλίδεω χρειώ με διηνεκέως ἀγορεύειν;

Aethalides has already been demonstrated to be a poetologically charged character,²² and now I would only take this argument one step further. As Nishimura-Jensen rightly points out, Apollonius omits the fact that Aethalides was said to have been an earlier incarnation of the soul that became Pythagoras himself. The words ἔτι νῦν περ (644), elsewhere markers of aetia, are not linked to a recognizable aetiological story and thus are left hanging. They should project the cycle of reincarnation from the mythical world of Aethalides into Apollonius’ present.²³

The reason for Apollonius’ sudden break-off is assumed to be his general avoidance of naming historical figures; but, strictly speaking, Pythagoras does not exactly fit the bill since by the time Apollonius was writing he would have been dead for quite some while, so it cannot be the end of the story.²⁴ Who else’s name can Apollonius be suppressing? Before attempting an alternative answer, I suggest that we consider an Empedoclean subtext Apollonius alludes to (fr. 115.6‒13): τρίς μιν μυρίας ὧρας ἀπὸ μακάρων ἀλάλησθαι, φυόμενον παντοῖα διὰ χρόνου εἴδεα θνητῶν ἀργαλέας βιότοιο μεταλλάσσοντα κελεύθους. αἰθέριον μὲν γάρ σφε μένος πόντονδε διώκει, πόντος δ’ ἐς χθονὸς οὖδας ἀπέπτυσε, γαῖα δ’ ἐς αὐγὰς ἠελίου φαέθοντος, ὁ δ’ αἰθέρος ἔμβαλε δίναις· ἄλλος δ’ ἐξ ἄλλου δέχεται, στυγέουσι δὲ πάντες. τῶν καὶ ἐγὼ νῦν εἰμι, φυγὰς θεόθεν καὶ ἀλήτης.

 Nishimura-Jensen (), though she argues that Aethalides represents a relatively marginal element of Apollonius’ epic poetics.  Nishimura-Jensen () ‒.  Though (μύθους) ἀγορεύειν does evoke Pythagoras’ name.

6.6

189

We already have seen this fragment used in Apollonius’ account of the Argo’s passage through the Wandering Rocks (2§8). In particular, I noted the enjambement ἐς αὐγὰς | ἠελίου, which we can see here imitated again, and with greater precision. Furthermore, the phrase ἔμπεδον αἰέν at 1.647 seems to be a specific Empedoclean ‘signpost’, as it also is in the cosmological song of Orpheus only a little earlier in the poem (1.499).²⁵ Finally, just as Empedocles, after describing what happens with the fallen δαίμονες in general, proceeds to claim that he, too, is one of them (τῶν καὶ ἐγὼ νῦν εἰμι), so, I would suggest, what is implied by Apollonius is that the present-day (ἔτι νῦν) incarnation of Aethalides’ soul is none other than the poet himself.²⁶ What does this mean for our appreciation of the figure of the narrator in the Argonautica? First of all, he is thus claimed to be an eyewitness of what he is narrating.²⁷ Moreover, he also turns out to be a competent judge of Homer, since one of Pythagoras’ prior (and Aethalides’ past) incarnations was as the Trojan Euphorbus killed by Menelaus (Il. 17.79‒81), a fact played upon by Ovid (Met. 15.160‒162).²⁸ These literal implications, however, are probably not as important as what may be inferred about Apollonius’ poetry on a metaphorical level. If Apollonius does indeed claim to possess an ἄφθιτον μνῆστιν πάντων, this memory should rather be interpreted as a thorough knowledge of poetic tradition, a knowledge that virtually equals every single author’s knowledge of his own works. It has been amply documented that in the Argonautica Apollonius develops a complex web of ‘window’ allusions as he systematically traces the poetic history of individual motifs or contexts (their subsequent reincarnations, as it were), for example through tragedy back to Homer:²⁹ this faculty can be said  On the latter context, cf. Hunter (b)  n.  and Kyriakou () ‒. The exact collocation ἔμπεδον αἰέν is not attested in Empedocles’ surviving fragments, but there are two instances of ἔμπεδος αἰών (fr. . and .). Cf. further Gale () ‒ on puns on Empedocles’ name in Latin poetry.  It may not be irrelevant to point out that Arg. .‒ ἄλλοθ’ ὑποχθονίοις ἐναρίθμιος, ἄλλοτ’ ἐς αὐγὰς | ἠελίου ζωοῖσι μετ’ ἀνδράσιν seems to be related to Theocr. . αἴθ’ ἐπ’ ἐμεῦ ζωοῖς ἐναρίθμιος ὤφελες ἦμεν, a context deeply concerned with matters of poetry (note also ἐπ’ ἐμεῦ – ἔτι νῦν). Add the parallel between Arg. .‒ ὃ δὴ κατὰ δῆμον ἄνασσε, | λάρνακι δ’ ἐν κοίλῃ μιν ὕπερθ’ ἁλὸς ἧκε φέρεσθαι and Theocr. .‒ ἔδεκτο τὸν αἰπόλον εὐρέα λάρναξ | ζωὸν ἐόντα κακαῖσιν ἀτασθαλίαισιν ἄνακτος.  Note that the so-called Orphic Argonautica has as its narrator Orpheus, i. e. another eyewitness.  On Ovid’s treatment see Miller () ‒.  Schmakeit () : “Apollonios’ intertextuelle Technik zeigt also nie eine eins-zu-eins Adaptation, sondern den Rückgriff auf mehrere Prätexte, die in der Regel miteinander zusammenhängen (two-tier Allusionen bzw. intertextuelle Verkettungen). Es ist auffällig, daß Apollonios oft einen homerischen Basistext mit Folgetexten aus der Tragödie aufnimmt, die darauf auf-

190

6 The metamorphosis (lines 478‒541)

to depend precisely on the fact that he himself is a reincarnation of the spirit – not only of Homer, but also of Aeschylus, Euripides, and so on.³⁰ Yet Apollonius’ imperishable and all-encompassing memory is different from that of an ordinary mortal not only quantitatively, but also, and this is a fundamental point, qualitatively: what is crucial, I suggest, is not so much his knowledge of a great many particular instances of a given motif per se, but rather his ability to grasp the ideal essence of that motif lurking behind the multitude of its actual occurrences. For a real understanding of particular phenomena must be grounded in knowledge of their underlying ontological archetype. This brings us back to what we discussed in the preceding section: in order to see the connection between two seemingly disparate uses of a given motif (which functions as a tertium comparationis, to return to the terminology we used there), the poet must know the ‘form’ of this motif, or in other words he needs an immortal soul that allows such knowledge. Such a Platonising poetics was, I suggest, a more or less standard way for Hellenistic poets to conceptualise their experience of writing poetry,³¹ and apparently it was rediscovered by Roman poets of the generation following Lucretius. What we find in the Ciris is, I believe, an attempt to reconcile this poetics – or, on the contrary, to make it clash? – with Epicureanism. For although Lucretius’ attack on the doctrine of metempsychosis was conducted within a philosophical rather than poetological framework, its implications were bound to affect poetics as well. To put the present inquiry in a broader context, we may look back at our earlier discussion of Aristotle’s statement that poetry is more philosophical than history (3§3). As we saw then, this implies that, in contrast to history, poetry narrates particular events, so to speak, sub specie aeternitatis, that is with respect to their abstract causal structure. This observation, I suggested, can be meaningfully extrapolated to refer not only to action, but also to include other, ‘static’, elements of poetic composition such as single motifs or even words: the poet, in gebaut haben. So spielt er auf Passagen und Motive aus Euripides, Aischylos und den homerischen Epen an, die miteinander zusammenhängen, wobei bereits Aischylos auf den homerischen Prätext aufbaut und Euripides seinerseits auf seinen älteren Tragikerkollegen anspielt.”  In fact, Apollonius can arguably be seen to do this precisely in our passage as he makes a ‘window’ allusion through the Empedoclean fragment to its well-known model in Hesiod’s Theogony (‒): cf. .‒ Αἰθαλίδην κήρυκα θοόν, τῷ πέρ τε μέλεσθαι | ἀγγελίας καὶ σκῆπτρον ἐπέτραπον Ἑρμείαο with ‒ παῦρα δὲ Θαύμαντος θυγάτηρ πόδας ὠκέα Ἶρις | ἀγγελίη πωλεῖται, .‒ ᾿Aχέροντος | δίνας ἀπροφάτους with  δίνῃς ἀργυρέῃς εἱλιγμένος εἰς ἅλα πίπτει (contrast the verse-end position of Emp. fr. . ὁ δ’ αἰθέρος ἔμβαλε δίναις).  Cf. Kayachev () ‒ on Callimachus.

6.7

191

contrast to the historian (or, more broadly, any sort of technical writer), will likewise use these elements with respect to their generalised abstract notion. Now we can easily see that it is precisely this conception that the figure of Aethalides in Apollonius, according to my reading, is intended to represent. If, then, the poet is concerned with universals (Aristotle’s τὰ καθόλου) and not just with individuals (τὰ καθ’ ἕκαστον), is it necessary that he take a Platonic, rather than Epicurean, stance on the problem of universals? It might perhaps be argued that Epicurean prolepses (general concepts produced by way of abstraction) could do as well as Platonic ideas (general concepts eternally existing in an ontological reality) to account for how poetry is composed. It is certainly too complex an issue to discuss here, but there is at least one point where Platonism has, as a poetological model, an undeniable intrinsic advantage. Universals of which the poem’s particular elements are individuals objectively (pre‐) exist in the author’s mind, and in this sense the poem’s universe is a Platonic one. Even from an Epicurean perspective, what are in the real world the author’s ‘prolepses’ become, when viewed from within the poem, ‘ideas’, existing in the ontological reality of the author’s mind.³²

6.7 What reasons are there to suppose that such a Platonising poetics is indeed present in the conceptual background of the Ciris? First of all, as I have been trying to show, especially in the chapter on the night episode, the complex and sophisticated manner in which the Ciris deals with its sources betrays, or at least suggests, a conscious application of precisely those compositional principles which this poetics was intended to conceptualise, namely what I described as the literary analogue of the oral-epic technique of composition by theme. Of course, this does not necessarily mean that the Ciris poet conceived of what he was doing in the same terms. Secondly, as we could observe on several occasions, love seems to have in the Ciris certain poetological implications, which I attempted to explicate, though without detailed argument, by positing a connection with the Platonic concept of love as the force that activates the process of recollection. In particular, I have argued that the statement of love’s omnipotence (437 omnia uicit amor) can be linked, through Callimachus’ hymn to Zeus and, possibly, Antagoras’ hymn to Eros, with Plato’s Symposium and specif-

 Cf. Sayers’s () account of the creative process through the analogy of the Christian understanding of creation.

192

6 The metamorphosis (lines 478‒541)

ically with the idea that love is the foundation of all manner of knowledge. And finally, the point we started with, the Ciris is demonstrably concerned with the issue of the (im)mortality of the soul, which in the Latin tradition would have had explicit poetological associations at least since Ennius. I propose that we now return to it and consider further relevant intertexts. Before adducing fresh material, however, I shall make some brief comments about two Empedoclean fragments we already have dealt with. The first relevant fragment is, of course, fr. 115; are there any signs that its use by Apollonius and, more generally, its poetological connotations were recognised by the Ciris poet? The evidence is tenuous, but it seems to point in that direction. On the one hand, as we have seen earlier, the Aethalides passage may be a subtext of the portrayal of Minos in Scylla’s speech, a context whose central subject-matter – Scylla’s love for Minos – I argued to have poetological implications. On the other, Empedocles’ fragment – with its combination of the motifs of crime, consisting in perjury, and punishment, consisting in metempsychosis, – provides an archetypal model for the Scylla story. Points of contact related to the former motif were noted in an earlier chapter (2§9); those related to the latter motif will be discussed later. The other fragment is fr. 128, which I suggested is alluded to – ‘through’ its earlier use in Lucretius – in the passage speaking of Nisus’ religiosity (4§3): as we remember, Empedocles’ reason for condemning animal sacrifices is nothing other than his doctrine of transmigration. Another case of ‘window’ reference can arguably be found in the allusion to Lucretius’ detailed description of the process of dying that he offers as proof of the soul’s mortality. As has been suggested, for this passage Lucretius is using, in his turn, Plato’s account of the death of Socrates in the Phaedo (117e–118a):³³ ὁ δὲ περιελθών, ἐπειδή οἱ βαρύνεσθαι ἔφη τὰ σκέλη, κατεκλίνη ὕπτιος – οὕτω γὰρ ἐκέλευεν ὁ ἄνθρωπος – καὶ ἅμα ἐφαπτόμενος αὐτοῦ οὗτος ὁ δοὺς τὸ φάρμακον, διαλιπὼν χρόνον ἐπεσκόπει τοὺς πόδας καὶ τὰ σκέλη, κἄπειτα σφόδρα πιέσας αὐτοῦ τὸν πόδα ἤρετο εἰ

 Kenney () , cf. De Lacy () ‒. The identification of the poison responsible for Socrates’ death has been a centuries-long problem, but Bloch () seems finally to have settled the issue by identifying the drug as poison hemlock (Conium maculatum). Its unique trait (among all other hemlocks) is that it affects the peripheral, rather than central, nervous system, which accounts for Socrates’ symptoms: ascending paralysis of the (lower) extremities with the mind remaining clear (cf. Aristophanes’ reference to hemlock poisoning, Ran. ‒ ψυχράν γε καὶ δυσχείμερον· | εὐθὺς γὰρ ἀποπήγνυσι τἀντικνήμια, with the scholium: ἀπὸ τῶν ποδῶν γὰρ οὗτος ὁ θάνατος ἄρχεται, πρώτους αὐτοὺς καταψύχων). It is difficult to decide whether the analogous symptoms of gradual paralysis described by Lucretius would have been likely to be observed often in other diseases, but the similarity with Plato, especially given the affinity of their arguments, can hardly be coincidental.

6.7

193

αἰσθάνοιτο, ὁ δ’ οὐκ ἔφη. καὶ μετὰ τοῦτο αὖθις τὰς κνήμας· καὶ ἐπανιὼν οὕτως ἡμῖν ἐπεδείκνυτο ὅτι ψύχοιτό τε καὶ πήγνυτο. καὶ αὐτὸς ἥπτετο καὶ εἶπεν ὅτι, ἐπειδὰν πρὸς τῇ καρδίᾳ γένηται αὐτῷ, τότε οἰχήσεται.

In particular we may point out that in both cases the process of dying begins from the feet and legs (pedes et crura, τοὺς πόδας καὶ τὰ σκέλη) and is manifested in a gradual loss of sensation (deperdere sensum, ἤρετο εἰ αἰσθάνοιτο, ὁ δ’ οὐκ ἔφη) and onset of coldness (gelidi uestigia leti, ἐπεδείκνυτο ὅτι ψύχοιτο). Additionally, both contexts employ the metaphor of ‘going’ twice each: once to describe the advance of death in the limbs (per artus | ire alios), or of the man tracing it (καὶ ἐπανιὼν οὕτως), and once as a euphemism for dying (hominem paulatim cernimus ire, τότε οἰχήσεται). Moreover, Lucretius clearly sees sensation as a function of the soul (ipsam se posse per artus | introsum trahere et partis conducere in unum | atque ideo cunctis sensum diducere membris), and it has persuasively been argued that the gradual loss of sensation experienced by Socrates is intended precisely to convey the idea of his soul being gradually freed from the body, a notion expressed explicitly earlier in the dialogue in terms clearly reminiscent of Lucretius’ phrasing (67c αὐτὴν καθ’ αὑτὴν πανταχόθεν ἐκ τοῦ σώματος συναγείρεσθαί τε καὶ ἁθροίζεσθαι, 83a αὐτὴν δὲ εἰς αὑτὴν συλλέγεσθαι καὶ ἁθροίζεσθαι).³⁴ Lucretius’ polemical point is that the soul leaves the body not undivided, as Plato believed, but in parts (dilaniata foras dispargitur, interit ergo). As De Lacy aptly comments, “to take from the Phaedo an argument for the mortality of the soul, an argument moreover drawn from the death of Socrates, the great champion of immortality, is a striking use of Plato against Plato.”³⁵ There are admittedly no specific points of contact between the Ciris and Plato which would attest to a direct connection, but two contexts in (arguably) later poetry alluding to the Ciris passage display, in their turn, demonstrable links with the Phaedo.

 Gill () : “[T]he physical details Plato does give […] can be seen as having a positive purpose, that of illustrating a major theme in the Phaedo: the liberation of the soul from the body. […]. The demonstration in which the poison-mixer pinches one part of the body after the other presents the spreading paralysis as the passing of sensation out of the body, proceeding stage by stage. […] [F]or Plato, sensation is an activity in which the psyche uses the body as its instrument. […] The gradual loss of sensation, then, would be seen as the departure of the psyche from the body […] – departure, in this case, having a special significance. […] Throughout the argument of the Phaedo, it is repeatedly emphasized that, in what is normally thought of as death, the psyche is not destroyed along with the body, but is released or purified from the body, and goes from it into independent, non-corporeal existence.”  De Lacy () ‒.

194

6 The metamorphosis (lines 478‒541)

6.8 The first context is the scene of Dido’s death in the Aeneid, which it will be helpful to quote at some length (4.693‒705): tum Iuno omnipotens longum miserata dolorem difficilisque obitus Irim demisit Olympo quae luctantem animam nexosque resolueret artus. nam quia nec fato merita nec morte peribat, sed misera ante diem subitoque accensa furore, nondum illi flauum Proserpina uertice crinem abstulerat Stygioque caput damnauerat Orco. ergo Iris croceis per caelum roscida pennis mille trahens uarios aduerso sole colores deuolat et supra caput astitit. ‘hunc ego Diti sacrum iussa fero teque isto corpore soluo’: sic ait et dextra crinem secat, omnis et una dilapsus calor atque in uentos uita recessit.

The connection with the Ciris may appear tenuous, but it must be stressed that Iris performs virtually the same act of cutting a lock of hair, thus causing the death of its owner, as Scylla did. Whether or not with the Ciris in mind (a point we shall return to shortly), Virgil seems to allude to the derivation of ciris from κείρειν, by evoking the former with Iris and the latter with crinem secat. ³⁶ At the same time, it has been suggested that, given the Platonic colouring of the underworld theme in the Aeneid in general,³⁷ Virgil’s account of Dido’s death may have as a subtext Plato’s treatment of the soul, in particular the following passage from the Phaedo (108a–b):³⁸ ἡ δ’ ἐπιθυμητικῶς τοῦ σώματος ἔχουσα [sc. ψυχή] … περὶ ἐκεῖνο πολὺν χρόνον ἐπτοημένη καὶ περὶ τὸν ὁρατὸν τόπον, πολλὰ ἀντιτείνασα καὶ πολλὰ παθοῦσα, βίᾳ καὶ μόγις ὑπὸ τοῦ προστεταγμένου δαίμονος οἴχεται ἀγομένη.

We may compare longum … dolorem with πολὺν χρόνον ἐπτοημένη and πολλὰ παθοῦσα, luctantem animam with πολλὰ ἀντιτείνασα and βίᾳ καὶ μόγις, animam nexosque resolueret artus and corpore soluo with ἐπιθυμητικῶς τοῦ σώματος ἔχουσα, Irim demisit with ὑπὸ τοῦ προστεταγμένου δαίμονος, and uita recessit with οἴχεται. Of course, even if Virgil links Dido’s death with the representation

 Cf. Konstan (), on a similar bilingual pun earlier in Aeneid .  On which see Michels (), West (), Weber ().  Weber () .

6.8

195

of Scylla’s metamorphosis in the Ciris, his simultaneous allusion to the Phaedo does not necessarily prove its relevance for the Ciris; but it certainly points in that direction. Now I shall offer some observations suggesting that it is the Ciris rather than any other (lost) treatment of the Scylla story that the figure of Iris is meant to evoke (which, needless to say, I do not intend as a conclusive demonstration of the Ciris’ anteriority). My argument is that one further Virgilian and one Ovidian context featuring Iris seem likewise to link her with the bird ciris, and specifically with the ciris’ appearance in our poem. In Aeneid 9 Iris is sent with a message for Turnus (9.14‒19): dixit, et in caelum paribus se sustulit alis ingentemque fuga secuit sub nubibus arcum. agnouit iuuenis duplicisque ad sidera palmas sustulit ac tali fugientem est uoce secutus: ‘Iri, decus caeli, quis te mihi nubibus actam detulit in terras?’

To begin with, we may point out secuit, a verb that we just have seen to evoke κείρειν;³⁹ moreover, the phrase fuga secuit seems specifically to recall the fourline passage depicting the perpetual pursuit of the ciris by the sea eagle, a passage shared verbatim by the Ciris (538‒541) and the Georgics (1.406‒409): quacumque illa leuem fugiens secat aethera pennis, ecce inimicus atrox magno stridore per auras insequitur Nisus; qua se fert Nisus ad auras, illa leuem fugiens raptim secat aethera pennis.

I would further suggest that the pattern of constant chasing and fleeing manifested in the behaviour of the two birds is alluded to by the phrase fugientem est uoce secutus. But there are also details that link the Aeneid passage with other contexts in the Ciris, not paralleled in the Georgics (or elsewhere). The rare verse-ending -tulit alis establishes a telling parallel with two interconnected passages of the Ciris (487‒489 and 514‒516):⁴⁰

 Though cf. Ap. Arg. .‒ αὐτίκα δ’ Ἶρις ἀπ’ Οὐλύμποιο θοροῦσα | τέμνε: Virgil may be interpreting Apollonius’ absolute use of τέμνειν.  In addition to the Virgilian context and the two in the Ciris, it is only attested once more in the Aeneid (in the doublet of the quoted passage at .‒), once in Ovid (Met. .), and once in Valerius Flaccus (.).

196

6 The metamorphosis (lines 478‒541)

aeriis potius sublimem sustulit alis, esset ut in terris facti de nomine ciris, ciris Amyclaeo formosior ansere Ledae. quae simul ut sese cano de gurgite uelox cum sonitu ad caelum stridentibus extulit alis et multum late dispersit in aequora rorem.

We can see that the Virgilian passage uses, like the former of the two, sustulit rather than extulit, but, like the latter (sese … extulit), couples it with a reflexive pronoun (se sustulit); similarly, in terras parallels in terris in the former, and in caelum parallels ad caelum in the latter (note that both phrases preserve their position in the line; moreover, all four occurrences refer to the location of either Iris or the ciris, in all four cases resulting from the action expressed by a -tulit verb). Furthermore, Virgil places Iri in the same metrical position as ciris: both contexts stress the beauty of either Iris or the ciris (cf. further decus caeli with Ciris 481 decus formae). Finally, I would suggest, ingentemque fuga secuit sub nubibus arcum can be linked with et multum late dispersit in aequora rorem: water spray, when refracting sunlight, produces a bow. Assuming the two Ciris contexts, on the one hand, and the Virgilian context, on the other, have a direct connection, are there any indications that the latter alludes to the former, not the other way round? Given the sophisticated character both of the Ciris, which I hope has become evident by now, and of Virgil’s poetry, which hardly needs any proof, it is difficult to find reliable arguments. We may try the following line of demonstration. As has been suggested earlier (5§3), sustulit at Ciris 487 is an allusion to Arg. 4.917‒918 ἀλλά μιν οἰκτείρασα θεὰ Ἔρυκος μεδέουσα | Κύπρις ἔτ’ ἐν δίναις ἀνερέψατο. Indeed, this subtext would also explain the only Ovidian use of -tulit alis in Met. 11.339‒341 miseratus Apollo … fecit auem et subitis pendentem sustulit alis: the bird metamorphosis parallels what happens in the Ciris (though Ovid’s hawk behaves more like the sea eagle than like the ciris: 11.344‒345 in omnes | saeuit aues), whereas miseratus is a precise equivalent of οἰκτείρασα. Moreover, the only instance of a cognate collocation sustulit ales at Aen. 5.861 (referring to Sleep) comes from a context that has among its models the Sirens episode in Apollonius to which the quoted passage belongs.⁴¹ How is this constellation of parallels to be accounted for? We may try to find an answer by asking a more specific question: which of the three texts is more likely to have introduced the connection with Apollonius? The Argonautica is, of course, a central model for the Aeneid, but Virgil’s sustulit does not have a

 See Nelis () ‒.

6.8

197

precise counterpart in Apollonius; unlike Ovid and the Ciris poet, who use sustulit transitively so as to match ἀνερέψατο, Virgil makes it a reflexive verb (se sustulit: in both 5.861 and 9.14). If now we are to choose between the Ciris and Ovid, perhaps the importance of the Argonautica as a model for the former can be decisive, though such a decision will depend on a negative judgement (that is, that Apollonius is not as important for the particular Ovidian context). Let us now turn to the Ovidian passage that can arguably be seen likewise to link Iris with the ciris (Met. 1.266‒271): barba grauis nimbis, canis fluit unda capillis, fronte sedent nebulae, rorant pennaeque sinusque. utque manu late pendentia nubila pressit, fit fragor; hinc densi funduntur ab aethere nimbi. nuntia Iunonis uarios induta colores concipit Iris aquas alimentaque nubibus adfert.

The portrayal of Iris as uarios induta colores does not need, of course, to be dependent on Ciris 502 uarios intexens pluma colores rather than Aen. 4.701 trahens uarios aduerso sole colores; but the figure of Notus, described in the preceding four lines, exhibits further notable points of resemblance with the ciris of our poem. Like Scylla during her metamorphosis, Notus combines features of both a bird and a human being. His whole portrait is grotesque, but perhaps its most comic detail is the picture of how he produces thunder and rain by squeezing the clouds like a sponge in his hand. In the Ciris, as we remember, Scylla’s arms shower a storm of feathers, if we take the text literally (504): perpetuas fuderunt bracchia pennas. As we also remember, the Ciris context alludes to Apollonius’ comparison of the Stymphalian birds shooting their feathers with a hailstorm sent by Zeus (2.1083‒1084, 1088): ὡς δ’ ὁπότε Κρονίδης πυκινὴν ἐφέηκε χάλαζαν | ἐκ νεφέων … ὣς πυκινὰ πτερὰ τοῖσιν ἐφίεσαν. Turning back now to the Ovidian passage, we may point out that funduntur is placed there in the same central – fourth-foot homodyne – position as fuderunt in the Ciris. Furthermore, the epithet densi matches perpetuas, both being renderings of, respectively, πυκινὴν (χάλαζαν) and πυκινὰ (πτερά).⁴² What we see in Ovid is, effectively, a literalisation of the Apollonian simile turned on its head: instead of comparing a shower of brazen feathers with a hailstorm, he assimilates Notus (who plays the role of Apollonius’ Zeus), as it were, to the Stymphalian birds; it is not even exactly clear whether the clouds are separate entities or, in fact, an integral part of

 As Bömer ()  points out, densus is not a standard epithet of rain.

198

6 The metamorphosis (lines 478‒541)

Notus’ person, just as feathers are part of a bird’s body (though easily detachable).⁴³ It is, again, extremely difficult to produce compelling arguments for the Ciris’ priority, especially as a possible shared model in Varro Atacinus’ Latin translation of the Argonautica could conceivably account for the little that the two contexts have in common in terms of precise textual correspondences. My main purpose, however, was to support the connection between Iris and the ciris, and if Ovid does allude here to the Ciris, it seems justified to assume that he does so with this connection in mind.

6.9 The other context that arguably combines parallels with the Ciris with references to Plato’s Phaedo (the first was Virgil’s account of Dido’s death) is Horace’s wellknown ‘swan ode’ 2.20 (1‒4, 9‒12): non usitata nec tenui ferar penna biformis per liquidum aethera uates, neque in terris morabor longius… … iam iam residunt cruribus asperae pelles, et album mutor in alitem superne, nascunturque leues per digitos umerosque plumae.

The ode’s Platonic backdrop has recently been discussed;⁴⁴ to give just one concise example, the content of the first stanza has been compared with Socrates’ assertion that (Phaed. 115d) οὐκέτι ὑμῖν παραμενῶ, ἀλλ’ οἰχήσομαι ἀπιὼν εἰς μακάρων δή τινας εὐδαιμονίας.⁴⁵ Now, if my proposed chronology is correct, Horace’s shrinking legs (residunt cruribus asperae | pelles) can be traced back, through the Ciris (506 crura noua macies obduxit squalida pelli)⁴⁶ and Lucretius

 We may recall P. Hardie’s ()  reference to “Ovid’s habit of generating tales of transformation out of a literalization of the figurative”.  Thévenaz () ‒, Erasmo () ‒, with further references. Cf. further Schiesaro () ‒ on Parmenidean and Pythagorean resonances in the ode.  Thévenaz () .  Lyne ()  notes the parallel between Horace and the Ciris; no other close parallels seem available.

6.9

199

(3.528‒529 in pedibus primum digitos liuescere et unguis, | inde pedes et crura mori), to the stiffening legs of Socrates (117e–118a ἐπεσκόπει τοὺς πόδας καὶ τὰ σκέλη … καὶ μετὰ τοῦτο αὖθις τὰς κνήμας· καὶ ἐπανιὼν οὕτως ἡμῖν ἐπεδείκνυτο ὅτι ψύχοιτό τε καὶ πήγνυτο).⁴⁷ What is even more important than providing additional support for the link between the Ciris and Plato, is, of course, Horace’s explicitly poetological use of a bird metamorphosis: if his swan is indeed modelled on the ciris, it obviously adds to the probability of the latter being a poetic bird as well. A further relevant context similarly speaking of the metamorphosis (or rather metempsychosis) of a poet into a bird – within an explicitly Pythagorean framework – is the closure of the pseudo-Tibullan Panegyricus Messallae, a poem known to feature a significant number of close textual correspondences with the Ciris (likewise addressed to Messalla)⁴⁸ (3.7.204‒211):⁴⁹ quin etiam mea tunc tumulus cum texerit ossa, seu matura dies celerem properat mihi mortem, longa manet seu uita, tamen, mutata figura seu me finget equum rigidos percurrere campos doctum seu tardi pecoris sim gloria taurus siue ego per liquidum uolucris uehar aera pennis, quandocumque hominem me longa receperit aetas, inceptis de te subtexam carmina chartis.

One such parallel is presented by the very last line of the panegyric, which seems to conflate (assuming it postdates the Ciris) 9 non tamen absistam coeptum detexere munus and 39 intexere chartis. It is therefore tempting also to link uehar aera pennis with the last line of the Ciris (541): secat aethera pennis. Even more remarkably, almost every word of 3.7.209 per liquidum uolucris uehar aera pennis finds a precise counterpart in the first two lines of Horace’s ode: non usitata nec tenui ferar | penna biformis per liquidum aethera (note that uehar and ferar are prosodically equivalent).⁵⁰ If the chronological sequence is indeed:

 Perhaps we can link in this connection iam iam with a ἤδη οὖν σχεδόν τι αὐτοῦ ἦν τὰ περὶ τὸ ἦτρον ψυχόμενα.  See Lyne () .  What the panegyric’s author speaks about is indeed a case of ἀλάλησθαι | φυόμενον παντοῖα διὰ χρόνου εἴδεα θνητῶν (Emp. fr. .‒). The notion of longa … aetas (cf. .. longior aetas) is reminiscent of Emp. fr. . ἄσπετος αἰών.  Noted by Peirano () . Tränkle ()  seems to be unaware of this Horatian parallel, but he mentions two others; Hollis () ‒ points out one more: “If the purported date of the Panegyricus Messallae ( BC) is indeed the true date, Horace would seem to have imitated these lines [i. e. ..‒] (by an unknown poet) in his Ode to Agrippa [..‒].”

200

6 The metamorphosis (lines 478‒541)

the Ciris – the panegyric – Horace,⁵¹ we can see aera and aethera replace one another in turn, starting with the Ciris’ model in Cicero’s Aratea (fr. 34.48 secat aera pennis). With the same proviso, I would tentatively suggest one further possible echo of the panegyric in Horace, though a less neutral one. Only a little before the quoted passage the poet tells Messalla that, for his sake, he would even Aetnaeae corpus committere flammae (3.7.196). We see Horace, as it were, take him at his word when at the end of the Ars poetica, after mentioning Empedocles who ardentem frigidus Aetnam | insiluit (465‒466),⁵² he advises (467): sit ius liceatque perire poetis. The panegyric’s author fits Horace’s portrayal of the uesanus poeta not only by the quality of his verse, but also by his intention never to stop writing about Messalla (3.7.203 nulla mihi statuent finem te fata canendi), not even after his death and a cycle of reincarnations: quandocumque hominem me longa receperit aetas, | inceptis de te subtexam carmina chartis (3.7.210‒211);⁵³ for Horace, in contrast, such a poet nec, si retractus erit, iam | fiet homo (468‒469), and is doomed to become – not a horse, or a bull, or a bird, but a non missura cutem nisi plena cruoris hirudo (476).⁵⁴ Perhaps we could also speculate that the Ciris, too, is criticised by Horace, this time in the opening passage of the Ars poetica (1‒4):⁵⁵

 If the Ciris is by Gallus and predates Virgil’s eclogues, it will be earlier than the panegyric and Horace’s ode. If the dramatic date of the panegyric is the real date, it will be earlier than Horace’s ode. Yet since Tränkle’s () ‒ argumentation for a later dating (in the first century AD), based on the analysis of textual parallels with other texts, scholars are reluctant to accept the genuineness of the panegyric, cf. recently De Luca () ‒ and Maltby (). This is not the place to offer a comprehensive discussion of the panegyric’s dating, but Tränkle’s arguments may not be entirely compelling, since they are mainly based on the (unprovable) assumption that a bad poem (which the panegyric is generally agreed to be) would not be alluded to by better poets (cf. my discussion of the dating of the Ciris in the Introduction). But even if the panegyric postdates both the Ciris and Horace, what we find in it could be interpreted as a ‘window’ allusion through Horace to the Ciris.  The parallel is pointed out by Peirano ()  n. .  He would also alios aliosque memor componere uersus (..), cf. Hor. Ars  nec satis apparet cur uersus factitet.  Of course, if the panegyric postdates Horace, a direct connection would be unlikely (though it may not be entirely inconceivable that the panegyric is a parody intended to fall under Horace’s criticism). Still it might be conjectured that both had a common model that would have depicted the poet as an Empedoclean figure, an idea I argue to be implicitly present in the Ciris.  Frischer () ‒ argued that Horace’s criticism of monsters in painting could not have been serious and concluded that the speaker of the Ars should be interpreted as a parody of a pedantic grammaticus. I have a certain sympathy for this (unpopular) view, but I rather believe Horace’s strategy to be more sophisticated: whilst on the surface level Horace’s instructions may

6.9

201

humano capiti ceruicem pictor equinam iungere si uelit et uarias inducere plumas undique collatis membris, ut turpiter atrum desinat in piscem mulier formosa superne…

Horace’s imaginary monster has naturally reminded many of mythological hybrid creatures, in particular of Scylla.⁵⁶ As we have seen, the Homeric Scylla is likewise lurking in the background of the Ciris’ last narrative part. A more direct (though hardly specific) point of contact is established by the monster’s motley plumage (uarias … plumas), a detail shared by the ciris (502 uarios intexens pluma colores).⁵⁷ The avian constituent of the Horatian monster may, however, be more prominent than appears at first sight, for Horace seems actually to evoke, among other things, the hybrid bird of ode 2.20.⁵⁸ It is probable that superne, noticeably avoided in poetry, may be intended to remind the reader in particular of that earlier use (2.20.10‒11 album mutor in alitem | superne).⁵⁹ Assuming this is enough for us to suspect, if not to prove, that the opening of the Ars poetica does indeed allude to the narrative of Scylla’s transformation, I would suggest that whereas in ode 2.20 Horace takes her as a figure for the poet, here she may stand, generally speaking, for the poem. As I have argued, in

sometimes seem banal, on the implicit level he wishes to show, by alluding to (and ridiculing) specific passages in earlier poetry (most of which are lost to us, since mainly masterpieces survive), how these rules can be applied in a more subtle way.  See e. g. Brink () .  Add perhaps ‒ internodia membris | imperfecta nouo fluitant and ‒ semiferi incertis etiam nunc partibus artus | undique mutabant atque undique mutabantur: the impression is that the ciris’ body is being shaped at random.  Cf. P. Hardie (a) : “At Odes . the poet’s posthumous metamorphosis into a swan produces something not unlike the hybrid monsters that exemplify an undisciplined poetica licentia at the beginning of the Ars poetica”; cf. (b) .  There are nineteen occurrences in Lucretius, and only one in Virgil (Aen. .). Catullus, Tibullus, and Propertius have none. Statius has three, Valerius Flaccus and Silius one each. Horace has three, the third being Sat. .. peccatue superne (on which see Wiesen [] ‒). It seems certain that Horace alludes to this third context in the Ars poetica as well. To start with, the next line (..) has mulier in the same metrical position as in Ars  (in both contexts, of course, it is a mulier who is superne). Further, .. iusta potestas can be linked with Ars  aequa potestas, and .. turpis,  turpi, with Ars  turpiter. If so, .. turgentis uerbera caudae may give an additional meaning to the fishtail at Ars  (desinat in piscem), and .. agitauit equum lasciua supinum may likewise expand the associations of the horse’s neck at Ars  (ceruicem pictor equinam). It is therefore hardly a chance coincidence that Ars  collatis membris evokes Lucr. . membris collatis (both collocations are unique). The result is thus an embodiment of promiscuity and autoeroticism paradoxically combined, cf. P. Hardie’s (b)  remarks on Aen. .‒. Cf. further Oliensis () ‒, ‒.

202

6 The metamorphosis (lines 478‒541)

the Ciris Scylla’s metamorphosis can be interpreted as a metaphor for allusion, and the somewhat grotesquely redundant character of the account of how it actually takes place (with an even greater number of phantasmagoric details present implicitly through allusion) may be meant to illustrate the uncontrollability of intertextual associations. Horace’s criticism of the unbridled imagination makes perfect sense when viewed from this perspective, too: a mulier formosa superne, that is on the narrative level of the poem, should not turn, as it were, into a hideous fish on the level of intertexts (a point Horace illustrates in this very passage by implicitly suggesting an obscene interpretation, see the preceding footnote).

6.10 Up until now I have been paying much attention to how the Ciris poet elicits the destabilising potential of thematic associations; now I shall try to demonstrate that he is likewise fully aware of their unifying force, which is a fundamental principle that underlies poetic composition. We may start, however, with a subtext that at first would appear, again, subversive: Apollonius’ account of Prometheus’ punishment by the monstrous eagle (2.1247‒1259): καὶ δὴ Καυκασίων ὀρέων ἀνέτελλον ἐρίπναι ἠλίβατοι, τόθι γυῖα περὶ στυφελοῖσι πάγοισιν ἰλλόμενος χαλκέῃσιν ἀλυκτοπέδῃσι Προμηθεὺς αἰετὸν ἥπατι φέρβε παλιμπετὲς ἀίσσοντα. τὸν μὲν ἐπ’ ἀκροτάτης ἴδον ἕσπερον ὀξέι ῥοίζῳ νηὸς ὑπερπτάμενον νεφέων σχεδόν, ἀλλὰ καὶ ἔμπης λαίφεα πάντ’ ἐτίναξε παραιθύξας πτερύγεσσιν· οὐ γὰρ ὅγ’ αἰθερίοιο φυὴν ἔχεν οἰωνοῖο, ἶσα δ’ ἐυξέστοις ὠκύπτερα πάλλεν ἐρετμοῖς. δηρὸν δ’ οὐ μετέπειτα πολύστονον ἄιον αὐδὴν ἧπαρ ἀνελκομένοιο Προμηθέος· ἔκτυπε δ’ αἰθὴρ οἰμωγῇ, μέσφ’ αὖτις ἀπ’ οὔρεος ἀίσσοντα αἰετὸν ὠμηστὴν αὐτὴν ὁδὸν εἰσενόησαν.

Although individual textual parallels with the Ciris may admittedly not be specific enough, a direct connection seems plausible. On the one hand, Scylla as the ciris can be linked with the eagle: both are said to produce a strident noise as they fly (515 stridentibus … alis, cf. ὀξέι ῥοίζῳ) and to cause violent perturbations in the air with their wings (516 multum late dispersit in aequora rorem, cf. λαίφεα πάντ’ ἐτίναξε παραιθύξας πτερύγεσσιν). In addition, the comparison of the eagle with a ship (ἶσα δ’ ἐυξέστοις ὠκύπτερα πάλλεν ἐρετμοῖς) is paral-

6.10

203

leled by the implicit identification of Scylla with the Argo which we have observed on several occasions. On the other hand, Scylla’s new habitat (519 rupibus et scopulis et litoribus desertis) strikingly resembles the place of Prometheus’ imprisonment (ἐρίπναι ἠλίβατοι and στυφελοῖσι πάγοισιν).⁶⁰ Moreover, just as Prometheus is punished by the eagle, the ciris is chased by the sea eagle (which likewise produces a strident noise: 539 magno stridore). So, on the surface, the effect of this allusion may be described as subversive (the beautiful ciris being again associated with a monster), or merely decorative at best. Yet on closer inspection, we can see that it has in fact a more specific point. The crucial trait is, I suggest, the regularity and repetitiveness of the eagle’s action, which Apollonius emphasises on the level of both content and form simultaneously: it is no coincidence that the phrasing of αἰετὸν ἥπατι φέρβε παλιμπετὲς ἀίσσοντα is recalled in αὖτις ἀπ’ οὔρεος ἀίσσοντα | αἰετὸν ὠμηστὴν αὐτὴν ὁδὸν εἰσενόησαν (note also παλιμπετές, αὖτις, αὐτὴν ὁδόν). We may at once point out the analogous (and even more conspicuous) symmetry in the closing quatrain of the Ciris, likewise intended to visualise the repetitiveness and regularity of the pursuit of the ciris by the sea eagle.⁶¹ Moreover, in Apollonius the patterned character of the eagle’s daily journey to the Caucasus and back can arguably be seen as providing an archetype for the Argonautic expedition itself: much as the eagle comes from the west to punish Prometheus, the Greek heroes go to the furthest east to retrieve their property. The comparison of the eagle to a ship is meant, I suggest, to point to this parallelism, which apparently had been recognised by Theocritus who, conversely, compares the Argo with an eagle (13.24 αἰετὸς ὥς).⁶² There is also one further potentially relevant implication: according to some accounts, it was precisely the eagle sent by Zeus to punish Prometheus that was later immortalised as the constellation Aquila; the regularity of the eagle’s flight observed by the Argonauts may therefore prefigure that of the course of a heavenly body. Here, too, we find a point of contact with the Ciris, for it explicitly compares the unceasing pursuit of the ciris by the sea eagle with that of Orion by Scorpius in the sky (533‒537). This possible astronomical undertone of the Apollonian allusion is corroborated and further developed by another potential subtext, rather unexpected.

 Note further the similarity with the Catullan reference to the same locality (. pendens e uerticibus praeruptis: note especially the spondaic ending).  It seems worth pointing out that just as in Apollonius the repeated phrase (αἰετὸν … ἀίσσοντα) is apparently a figura etymologica (cf. e. g. Et. magn., s.v. ἀετός: παρὰ τὸ ἀίσσω, τὸ ὁρμῶ, ἀιτὸς καὶ ἀετός), so in the Ciris the repeated secat aethera pennis alludes to the derivation of ciris from κείρειν.  On the relative chronology of Apollonius and Theocritus, cf. Hunter () ‒.

204

6 The metamorphosis (lines 478‒541)

The only extant fragment of Moero’s Mnemosyne tells an alternative aetiological story about Aquila (fr. 1.5‒8):⁶³ νέκταρ δ’ ἐκ πέτρης μέγας αἰετὸς αἰὲν ἀφύσσων γαμφηλῇς φορέεσκε ποτὸν Διὶ μητιόεντι. τῷ καὶ νικήσας πατέρα Κρόνον εὐρύοπα Ζεὺς ἀθάνατον ποίησε καὶ οὐρανῷ ἐγκατένασσεν.

The same pattern of merit and reward is present in the account of Nisus’ transformation into the sea eagle: it was because Nisus had regularly performed sacrifices (524‒526) that Jupiter (527‒529) reddidit optatam mutato corpore uitam fecitque in terris haliaeetos ales ut esset: quippe aquilis semper gaudet deus ille coruscis.

To begin with a formal observation, (hali)aeetos is placed in the same metrical position as αἰετός.⁶⁴ As we just have seen, Apollonius apparently puns on the derivation of αἰετός from ἀίσσειν, but Moero seems to play on an alternative etymology from αἰεί (and ἔτος), which in the Ciris is arguably hinted at by the collocation aquilis semper (‘translating’ αἰετὸς αἰέν).⁶⁵ In fact, the etymological implications of the Ciris passage may be even more complex since aquila – traditionally linked with aqua ⁶⁶ – would be a particularly suitable equivalent for the sea eagle (haliaeetos, from ἅλς).⁶⁷ The presence of the celestial Eagle in the background of the Ciris may additionally be confirmed by the fact that secat aethera pennis is a conflation of Cicero’s (fr. 34.48) secat aera pennis, referring to Cygnus, and (88) mulcens tremebundis aethera pennis, referring to Aquila.⁶⁸

 For a recent reappraisal of this half-forgotten Hellenistic poetess and in particular of her epic Mnemosyne, see Skinner () ‒.  Likewise, in each case in the next line, deus is placed in the same position as Διί: this type of intertextual wordplay is paralleled, for example, in Verg. Georg. . in primis uenerare deos, atque annua magnae… rendering Hes. Op.  εὔχεσθαι δὲ Διὶ χθονίῳ Δημήτερί θ’ ἁγνῇ…  Cf. Et. Gud., s.v. ἀετός: ἀείετος, ὁ ἀεὶ νέος ὢν διὰ τὸ πολλάκις ἀνακαινίζεσθαι καὶ πολλὰ 〈ἔτη〉 ζῆν. Possibly the same etymology is present in ‒ sic inter sese tristis haliaeetos iras | et ciris memori seruant ad saecula fato, with saecula singling out ἔτος rather than ἀεί.  See Maltby () .  Note that Cicero seems to imitate Aratus’ wordplay on ᾿Aητόν () and ἄηται (), on which see Kidd () , with Aquila (fr. .) and Aquilonis ().  Cf. Salvatore () ‒. Note Catullus’ earlier conflation of the same two Ciceronian contexts in . impellens nutantibus aera pennis.

6.11

205

It is true that what the two Greek subtexts, as well as the explicitly astronomical simile in 533‒537, link to an eternal celestial paradigm is not, in the first place, the entire Scylla story, but specifically its outcome. However, the astronomical dimension is perceptibly present in the Ciris from its very beginning, which is modelled on Catullus 65 introducing a translation of Callimachus’ aetiological account of the constellation Coma Berenices (1§2). Likewise, although explicit reference is only made to the constellations Orion and Scorpius, the story leading to their catasterism is arguably alluded to throughout the poem: 154 uiolata manu sacraria diuae may contain an echo of Cic. Arat. fr. 34.420 Orion manibus uiolasse Dianam; 163‒164 quae simul ut uenis hausit sitientibus ignem | et ualidum penitus concepit in ossa furorem, that of fr. 34.431– 432 hic ualido cupide uenantem perculit ictu, | mortiferum in uenas figens per uulnera uirus. ⁶⁹ We may also note two pointed internal reminiscences. First, 7 altius ad magni suspexit sidera mundi (modelled on Catul. 66.1 omnia qui magni dispexit lumina mundi) is echoed in 217‒218 alte | suspicit ad celsi nictantia sidera mundi, thus linking Scylla with the narrator as an ‘astronomer’. Second, 37‒38 purpureos inter soles et candida lunae | sidera, caeruleis orbem pulsantia bigis (referring to a didactic poem on the universe) provides a model for 102‒103 Actaeos inter colles et candida Thesei | purpureis late ridentia litora conchis (which creates the setting for the main narrative). The latter pair of parallels is especially significant as it undermines the narrator’s ironic rejection of a didactic poem in favour of a mythological one by showing that his account of a particular earthly event will be patterned, in a sense, after the model of the eternal and universal laws followed by the heavenly bodies (cf. above 1§7).

6.11 Metamorphosis, it has been suggested, can “be understood as primarily a process of abstraction.”⁷⁰ We see this take place as the pursuit of the ciris by the sea eagle perpetuates Nisus’ odium crudele (532) for his daughter, which, in a sense, characterised their relationship from the very beginning. Even more generally, anger is both what sets the whole story in motion (138‒139 idem tum tristis acue-

 As Lyne () , commenting on , plausibly argues, “what is influencing our line directly is Cicero’s version of the pursuit – unfortunately lost in a lacuna at Arat. .” On the use of the Aratea elsewhere in the Ciris, cf. Lyne () ‒ and Salvatore (). Kubiak () offers interesting comments on the Orion episode, showing that Cicero reworks Aratus’ original in such a way as to emphasise the potential erotic resonances of the story.  P. Hardie () , following Solodow () ‒.

206

6 The metamorphosis (lines 478‒541)

bat paruulus iras | Iunonis magnae) and what is its everlasting outcome (536‒537 sic inter sese tristis haliaeetos iras | et ciris memori seruant ad saecula fato). This is another charged internal echo, intended to convey, I suggest, the idea of strife as the ultimate principle of existence. Juno’s anger is caused by a combination of desecration and perjury, a crime reminiscent, as we have seen (2§9), of the crime for which the Empedoclean δαίμονες are punished with a cycle of reincarnations (fr. 115.3‒4). Now we can see that Scylla’s punishment, too, is not unlike theirs, for much as they are ‘chased’ (115.9 διώκει) and ‘hated’ (115.12 στυγέουσι δὲ πάντες) by all the elements in turn, so the ciris eternally flees from the sea eagle’s hatred (538‒541):⁷¹ quacumque illa leuem fugiens secat aethera pennis, ecce inimicus atrox magno stridore per auras insequitur Nisus; qua se fert Nisus ad auras, illa leuem fugiens raptim secat aethera pennis.

An even more radical reduction is effected by a possible Lucretian subtext which argues that everything can ultimately be analysed into two categories only, matter and void (1.507‒509): nam quacumque uacat spatium, quod inane uocamus, corpus ea non est; qua porro cumque tenet se corpus, ea uacuum nequaquam constat inane.

Lucretius goes even further as he states, only a few lines earlier, that all events, represented by the particular case of the Trojan war, are merely accidents – a function – of the interaction of matter and void (1.471‒477).⁷² Such a striking reduction of the poem’s entire plot to the mere interaction of matter and void can probably be viewed as a radical application of Aristotle’s statement, which we discussed earlier, that φιλοσοφώτερον καὶ σπουδαιότερον ποίησις ἱστορίας  Scylla’s seclusion from mankind ( numquam illam post haec oculi uidere suorum) can be compared with that of the fallen δαίμονες (. τρίς μιν μυρίας ὧρας ἀπὸ μακάρων ἀλάλησθαι). Cf. further Ov. Met. . tellusque tibi pontusque negetur with . πόντος δ’ ἐς χθονὸς οὖδας ἀπέπτυσε, γαῖα δ’ ἐς αὐγάς…  The same wider context seems also to be alluded to in Scylla’s monologue when she complains (‒): non equidem ex isto speraui corpore posse | tale malum nasci, which evokes Lucr. .‒ difficile esse uidetur credere quicquam | in rebus solido reperiri corpore posse (the collocation corpore posse is not attested elsewhere). Generally speaking, both contexts play with the idea that, against expectations (non … speraui; difficile … credere), what is produced (tale malum; composite bodies) may be in some respect unlike what it is produced from (ex isto … corpore; atoms).

6.11

207

ἐστίν· ἡ μὲν γὰρ ποίησις μᾶλλον τὰ καθόλου, ἡ δ’ ἱστορία τὰ καθ’ ἕκαστον λέγει (1451b). The narrative of the Ciris can thus be said to be patterned, as it were, on the matrix of the universal principles (τὰ καθόλου) of nature. If we ask how this mechanism manifests itself on the level of allusion, we can see that the fundamental ‘irreconcilability’ of matter and void – the most abstract, though still visual, expression of the idea of ‘strife’ – functions as, or perhaps is an image of, the conceptual axis around which a number of intertexts conveying the same ‘basic idea’ are organised. The hostility of the Stymphalian birds towards the Argo, that of Zeus’ eagle towards Prometheus, that of the four elements towards the fallen δαίμονες – all these examples that underlie the hostility of Nisus to Scylla – are, so to speak, subsumed under that most elemental ‘conflict’ between void and matter. To use the metaphor again, this is the soul of that particular group of allusions; the paradox is that this ‘soul’ is represented by the core materialist dogma that there is nothing but matter and void. As a final touch, let us address this fundamental paradox, created by the attempt to reconcile a Platonising poetics with Epicureanism, once more. Metamorphosis, as we have seen, is a very convenient image to visualise the unity of any two seemingly disparate, but implicitly related, levels of a poetic text, whether it be the tenor and the vehicle of a metaphor, or the explicit surface meaning and the meaning implied by a subtext. In figurative terms, for a metaphor (or allusion) to take place, there must be an immortal soul that would connect the two parts of that metaphor (or allusion). Yet, as we have had several occasions to note, the Ciris poet seems to undermine this model by alluding to Lucretius’ argument against the immortality of the soul. It is one more such subtext that I would consider in conclusion (3.748‒756):⁷³ quod si inmortalis foret et mutare soleret corpora, permixtis animantes moribus essent, effugeret canis Hyrcano de semine saepe cornigeri incursum cerui tremeretque per auras aeris accipiter fugiens ueniente columba, desiperent homines, saperent fera saecla ferarum. illud enim falsa fertur ratione, quod aiunt inmortalem animam mutato corpore flecti; quod mutatur enim, dissoluitur, interit ergo.

 To make a formal argument, the collocation mutato corpore, as already pointed out, has only three occurrences in Latin verse, which virtually proves a direct dependence of Ciris  reddidit optatam mutato corpore uitam on this Lucretian context (the third being Verg. Georg. .).

208

6 The metamorphosis (lines 478‒541)

Lucretius is arguing here a contrario: if souls did not die with their bodies and could migrate into new ones, it should happen from time to time that, for example, a hawk’s soul would be reborn in a dove’s body (and the other way round), so that we might occasionally see a dove chasing a hawk; since we never see that, it must be concluded that souls die with their bodies. Turning now to the Ciris, we realise that, in the light of the Lucretian subtext, the constant pursuit of the ciris by the sea eagle constitutes in itself a proof that the two bird species cannot have originated as reincarnations of Scylla and Nisus. But in reality the paradox is even deeper. If we take metamorphosis as an image for figurative language, that is if we interpret body and soul as form and content respectively, we can see that the Lucretian allusion refutes its own implication. For allusion only takes place when the reader transfers the content of one context into another, or in other words when the meaning of the model is reincarnated in the imitation. Moreover, what we witness happen in the Ciris is precisely a case of ‘hawk-todove’ metempsychosis, since what on the surface seems to be a Pythagorean text turns out to be imbued, by means of allusion, with an Epicurean meaning.

Conclusion The primary subject of exploration in this study has been the use of pre-Virgilian literary models in the Ciris. Some of these models, of course, were known before, but no systematic treatment of them existed. Although it may often be difficult to decide on their relative importance for the Ciris, partly because some texts only survive in fragments, an approximate arrangement can be made. It seems clear that, among extant texts, Catullus 64, Lucretius and Apollonius’ Argonautica were principal models, and it seems likely that Callimachus’ Hecale, had it come down to us complete, would have fallen within the same category. Yet there are, it should be noted, perceptible differences between the uses of, say, Lucretius and Apollonius, due to the fact that they belong to different genres and different languages. Texts that seem to be of a somewhat more episodic relevance are Catullus 65‒66, Callimachus’ Aetia (partly through the intermediary of Catullus) and hymns to Artemis and Delos, Moschus’ Europa (often because of its thematic and genetic affinities with Apollonius), and Empedocles (likewise, often because of his connections with Lucretius). The Iliad, Euripides’ Hippolytus, Pindar’s fourth Pythian ode, and Cicero’s Aratea could perhaps be put in the same group. Finally, individual allusions to the following texts have been noted, in some cases only tentatively: Catullus 68 (with its subtext in Philodemus), the Rhetoric to Herennius, the Odyssey, the Homeric hymn to Demeter, Callimachus’ hymn to Zeus, some of Theocritus’ idylls, Nicander’s Theriaca and Alexipharmaca; (the fragments of) the Cypria, Euphorion’s Thrax and Hyacinthus, Bion, Choerilus of Samos, Asius of Samos, Moero, Ennius’ Medea and Annals, and Cicero’s De consulatu suo. While the suggested parallels with some of these texts may conceivably be coincidental, it is also probable that the use of others, especially those known only in fragments, would have been significantly more extensive. Perhaps the most characteristic feature of the Ciris’ allusive technique is the frequent, even systematic, employment of what are called ‘window’ references. For example, the allusion to Catullus 65 at the opening of the Ciris is apparently accompanied by an allusion to Catullus’ own model in the exordium of the Rhetoric to Herennius. Likewise, the allusion to the catalogue of flowers in Moschus’ Europa at the end of the proem seems also to acknowledge Moschus’ use of analogous contexts in the Homeric hymn to Demeter and in the Cypria. In an even more complex way, the night episode as a whole can be said to be modelled on the night scene in the Europa, the night episode in Apollonius’ Argonautica and the Homeric Doloneia, which, in turn, are genetically linked as Apollonius alludes to Homer and Moschus alludes to both of them. It seems probable

210

Conclusion

that if more texts had survived, and if the interrelations between the Ciris’ extant models had been studied more extensively, the resulting picture would be even more intricate. It has also been argued that on an implicit level, partly created by allusions to this multitude of subtexts, the Ciris develops an allegorical ‘narrative’ with a certain poetological message, though, of course, it could only be recovered in fragments and tentatively. What happens within this allegorical dimension can perhaps be described, retrospectively, as a thought experiment devised to explore the collision of a materialist (‘Epicurean’ / ‘Lucretian’) and a metaphysical (‘Platonic’ / ‘Empedoclean’) poetics through the figure of the protagonist embodying the metaphor of the poet as lover. Thus, Scylla’s love was characterised, by means of allusion, both as a purely physiological condition and as an aspiration for the transcendent. A similar, and perhaps even more crucial, dichotomy concerned the nature of the soul: one group of subtexts implied that it is mortal, whilst the other that it is immortal. This latter dichotomy is the clearest illustration of the most fundamental issue of poetics: what is more elementary, words (form) or meanings (content)? In more historical terms, it was suggested that this tension between two conflicting poetics could be seen to arise from a collision of the poetic theories of Lucretius and Philodemus on the one hand and certain fundamental trends in Hellenistic poetry on the other, a collision that would have been particularly natural in the Late Republican period. Although establishing the Ciris’ date of composition was not among the primary goals, the character of the poem’s use of pre-Virgilian models strongly suggests that a date before Virgil’s Eclogues, and as a consequence the attribution to Gallus, are worth serious consideration. On the one hand, both the quantity and the quality of the Ciris’ engagement with (especially) Greek texts points at a wellread poet as its author. On the other, the Ciris’ poetological concerns manifest considerable influence from Lucretius and Philodemus, who had an immense impact on the poets of Virgil’s generation. Furthermore, both these facts also attest to the seriousness of purpose inherent in the Ciris, which seems inconsistent with the view of the poem as either an artless cento-like compilation or a sophisticated literary game. The author of the present study is only too well aware of its provisional character as a holistic reading of the Ciris. No doubt, new models can be identified, as well as further parallels with known models, and more subtle and detailed interpretations can be offered. Yet perhaps the most fruitful field of inquiry for future research could be the Ciris’ relation to Augustan poets, not only Virgil, but also Ovid, Horace, Propertius, and others, in the light of its engagement with earlier texts. Several separate cases of a possible ‘window’ allusion by an arguably later author through the Ciris to its models (or, if assuming a later

Conclusion

211

date of composition, by the Ciris through an Augustan poet to his models) have been treated in this study, but a more systematic approach, taking into account extended narrative segments rather than isolated small-scale contexts, is likely to produce a deeper and perhaps clearer understanding of the Ciris’ place within the Roman poetic tradition.

Bibliography Titles of periodicals are abbreviated according to the list of L’Année philologique. Abbreviated references to ancient authors and texts normally follow the conventions of LSJ (H.G. Liddell, R. Scott, H.S. Jones, A Greek-English Lexicon, rev. ed., Oxford 1996) and OLD (P.G.W. Glare, Oxford Latin Dictionary, Oxford 1968‒1982), or are more explicit. TLL is Thesaurus linguae Latinae, Leipzig 1894‒; LfgrE is Lexikon des frühgriechischen Epos, Göttingen 1955‒2010. SH is H. Lloyd-Jones, P. Parsons, Supplementum Hellenisticum, Berlin 1983. Acosta-Hughes and Stephens (2012): B. Acosta-Hughes, S.A. Stephens, Callimachus in Context: From Plato to the Augustan Poets, Cambridge. Aicher (1989): P. Aicher, “Ennius’ dream of Homer”, AJPh 110, 227‒232. Albis (1996): R.V. Albis, Poet and Audience in the Argonautica of Apollonius, Lanham. Ambühl (2005): A. Ambühl, Kinder und junge Helden: Innovative Aspekte des Umgangs mit der literarischen Tradition bei Kallimachos, Leuven. Ardizzoni (1965): A. Ardizzoni, “Echi pitagorici in Apollonio Rodio e Callimaco”, RFIC 93, 257‒ 267. Arend (1933): W. Arend, Die typischen Scenen bei Homer, Berlin. Armstrong (1995): D. Armstrong, “The impossibility of metathesis: Philodemus and Lucretius on form and content in poetry”, in: D. Obbink (ed.), Philodemus and Poetry: Poetic Theory and Practice in Lucretius, Philodemus, and Horace, New York, 210‒232. Armstrong (1998): J.M. Armstrong, “Aristotle on the philosophical nature of poetry”, CQ 48, 447‒455. Armstrong (2006): R. Armstrong, Cretan Women: Pasiphae, Ariadne, and Phaedra in Latin Poetry, Oxford. Asheri (2007): D. Asheri, “Book I”, in: D. Asheri, A. Lloyd, A. Corcella, A Commentary on Herodotus Books I–IV, Oxford, 58‒218. Axelson (1960): B. Axelson, “Lygdamus und Ovid: Zur Methodik der literarischen Prioritätsbestimmung”, Eranos 58, 92‒111. Bär (2012a): S. Bär, “War die Argo nicht doch das erste Schiff?”, RhM 155, 210‒214. Bär (2012b): S. Bär, “Zivilisationskritik und Kulturpessimismus im Ammenprolog von Euripides’ Medea”, Gymnasium 119, 215‒233. Barchiesi (1994): A. Barchiesi, “Immovable Delos: Aeneid 3.73‒98 and the hymns of Callimachus”, CQ 44, 438‒443. Barchiesi (2005): A. Barchiesi, “The search for the perfect book: a PS to the New Posidippus”, in: K. Gutzwiller (ed.), The New Posidippus: A Hellenistic Poetry Book, Oxford, 320‒342. Bartels (2004): A. Bartels, Vergleichende Studien zur Erzählkunst des römischen Epyllion, Göttingen. Belling (1909): H. Belling, “Die Ciris”, JPhV 35, 146‒184. Bellinger (1922): A.R. Bellinger, “Catullus and the Ciris”, TAPhA 53, 73‒82. Bloch (2002): E. Bloch, “Hemlock poisoning and the death of Socrates: did Plato tell the truth?”, in: T.C. Brickhouse, N.D. Smith (eds.), The Trial and Execution of Socrates: Sources and Controversies, New York, 255‒278. Bollack (1969): J. Bollack, Empédocle III: Les Origines, commentaire 1, Paris. Bömer (1969): F. Bömer, P. Ovidius Naso: Metamorphosen, Buch I–III, Heidelberg.

Bibliography

213

Bornmann (1968): F. Bornmann, Callimachi hymnus in Dianam: introduzione, testo critico e commento, Firenze. Braswell (1988): B.K. Braswell, A Commentary on the Fourth Pythian Ode of Pindar, Berlin. Breed (2000): B.W. Breed, “Imitations of originality: Theocritus and Lucretius at the start of the Eclogues”, Vergilius 46, 3‒20. Bretzigheimer (2005): G. Bretzigheimer, “Poeta memor ludensque oder The Making of Ciris”, in: N. Holzberg (ed.), Die Appendix Vergiliana: Pseudepigraphen im literarischen Kontext, Tübingen, 142‒224. Brink (1971): C.O. Brink, Horace on Poetry: The ‘Ars Poetica’, Cambridge. Brown (1982): R.D. Brown, “Lucretius and Callimachus”, ICS 7, 77‒97. Brown (1987): R.D. Brown, Lucretius on Love and Sex: A Commentary on De Rerum Natura IV, 1030‒1287, Leiden. Bühler (1960): W. Bühler, Die Europa des Moschos: Text, Übersetzung und Kommentar, Wiesbaden. Bulloch (2006): A.W. Bulloch, “Jason’s cloak”, Hermes 134, 44‒68. Burkert (1969): W. Burkert, “Das Proömium des Parmenides und die Katabasis des Pythagoras”, Phronesis 14, 1‒30. Burrow (2008): C. Burrow, “English Renaissance readers and the Appendix Vergiliana”, PVS 26, 1‒16. Butrica (2007): J.L. Butrica, “History and transmission of the text”, in: M.B. Skinner (ed.), A Companion to Catullus, Malden, MA, 13‒34. Cairns (1972): F. Cairns, Generic Composition in Greek and Roman Poetry, Edinburgh. Cairns (1996): F. Cairns, “Ancient ‘etymology’ and Tibullus: on the classification of ‘etymologies’ and on ‘etymological markers’ ”, PCPhS 42, 24‒59. Cairns (2011): F. Cairns, “Philodemus AP 5.123, the epigrammatic tradition, and Propertius 1.3”, in: A. Keith (ed.), Latin Elegy and Hellenistic Epigram: A Tale of Two Genres at Rome, Newcastle, 33‒50. Cameron (1995): A. Cameron, Callimachus and His Critics, Princeton. Campbell (1983): M. Campbell, Studies in the Third Book of Apollonius Rhodius’ Argonautica, Hildesheim. Campbell (1991): M. Campbell, Moschus: Europa, Hildesheim. Campbell (1994): M. Campbell, A Commentary on Apollonius Rhodius, Argonautica III 1‒471, Leiden. Campbell (2008): G. Campbell, “ ‘And bright was the flame of their friendship’ (Empedocles B130): humans, animals, justice, and friendship, in Lucretius and Empedocles”, LICS 7.4, 1‒23. Carawan (2000): E. Carawan, “Deianira’s guilt”, TAPhA 130, 189‒237. Casali (2004): S. Casali, “Nisus and Euryalus: exploiting the contradictions in Virgil’s Doloneia”, HSCPh 102, 319‒354. Caston (2006): R.R. Caston, “Love as illness: poets and philosophers on romantic love’, CJ 101, 271‒298. Cazzaniga (1960): I. Cazzaniga, “Osservazioni critiche intorno ad alcuni passi della Ciris”, SIFC 32, 125‒145. Cerri (2004): G. Cerri, “Empedocle, fr. 3 D.-K.: saggio di esegesi letterale”, in: L. Rossetti, C. Santaniello (eds.), Studi sul pensiero e sulla lingua di Empedocle, Bari, 83‒94. Clare (1996): R.J. Clare, “Catullus 64 and the Argonautica of Apollonius Rhodius: allusion and exemplarity”, PCPhS 42, 60‒88.

214

Bibliography

Clare (2002): R.J. Clare, The Path of the Argo: Language, Imagery and Narrative in the Argonautica of Apollonius Rhodius, Cambridge. Clausen (1964): W. Clausen, “Callimachus and Latin poetry”, GRBS 5, 181‒196. Clauss (1993): J.J. Clauss, The Best of the Argonauts: The Redefinition of the Epic Hero in Book 1 of Apollonius’s Argonautica, Berkeley. Colace (1979): P. Radici Colace, Choerili Samii Reliquiae, Roma. Connors (1991): C. Connors, “Simultaneous hunting and herding at Ciris 297‒300”, CQ 41, 556‒559. Conte (1986): G.B. Conte, The Rhetoric of Imitation: Genre and Poetic Memory in Virgil and Other Latin Poets, Ithaca. Cook (1940): A.B. Cook, Zeus: A Study in Ancient Religion, Vol. III, Part I, Cambridge. Cuypers (2004): M. Cuypers, “Prince and principle: the philosophy of Callimachus’ Hymn to Zeus”, in: M.A. Harder, R.F. Regtuit, G.C. Wakker (eds.), Callimachus II, Leuven, 95‒155. Dal Zotto (1903): A. Dal Zotto, La Ciris e le sue fonti greche, Feltre. Davies (1991): M. Davies, Sophocles: Trachiniae, Oxford. De Gianni (2010): D. De Gianni, “La nutrice di Scilla e la nutrice di Fedra: ispirazioni euripidee nella Ciris”, Vichiana 12, 36‒45. De Jonge (1995): I.J.F. De Jonge, “Homer as literature: some current areas of research”, in: J.P. Crielaard (ed.), Homeric Questions: Essays in Philology, Ancient History, and Archaeology, Amsterdam, 127‒146. De Jonge (2005): C.C. De Jonge, “Dionysus of Halicarnassus and the method of metathesis”, CQ 55 (2005) 463‒480. De Lacy (1983): P.H. De Lacy, “Lucretius and Plato”, in: Συζήτησις: Studi sull’epicureismo greco e romano offerti a Marcello Gigante, Napoli, 291‒307. De Luca (2009): E. De Luca, Corpus Tibullianum III 7: Panegyricus Messallae, Soveria Mannelli. Depew (2004): M. Depew, “Gender, power, and poetics in Callimachus’ book of Hymns”, in: M.A. Harder, R.F. Regtuit, G.C. Wakker (eds.), Callimachus II, Leuven, 117‒137. Dickie (1998): M. Dickie, “Poets as initiates in the mysteries: Euphorion, Phillicus and Posidippus”, A&A 44, 49‒77. Dirlmeier (1966): F. Dirlmeier, Die Giftpfeile des Odysseus (Zu Odyssee 1, 252‒266), Heidelberg. Doeblin (1968): R. Doeblin, [Review of W.V. Clausen et al., Appendix Vergiliana, Oxford 1966], CPh 63, 247‒251. Drachmann (1908): A.B. Drachmann, “Zur Cirisfrage”, Hermes 43, 405‒426. Duckworth (1966): G.E. Duckworth, “Studies in Latin hexameter poetry’, TAPhA 97, 67‒113. Dué and Ebbot (2010): C. Dué, M. Ebbot, Iliad 10 and the Poetics of Ambush: A Multitext Edition with Essays and Commentary, Cambridge, MA. Edgeworth (1979): R.J. Edgeworth, “Does ‘purpureus’ mean ‘bright’?”, Glotta 57, 281‒291. Edgeworth (1987): R.J. Edgeworth, “ ‘Off-color’ allusions in Roman poetry”, Glotta 65, 134‒137. Edmunds (2001): L. Edmunds, Intertextuality and the Reading of Roman Poetry, Baltimore. Edmunds (2002): L. Edmunds, “Mars as Hellenistic lover: Lucretius, De rerum natura 1.29‒40 and its subtexts”, IJCT 8, 343‒358. Edwards (1977): S.R. Edwards, “The Narcissus myth in Spenser’s poetry”, SPh 74, 63‒88. Edwards (1990): M.J. Edwards, “Treading the aether: Lucretius, De Rerum Natura 1.62‒79”, CQ 40, 465‒469. Edwards (1992): M. W. Edwards, “Homer and oral tradition: the type-scene”, Oral Tradition 7, 284‒330.

Bibliography

215

Eglinus (1606): R. Eglinus, “Vindiciae Ciris Catullianae”, in: id., C. Valerii Catulli casta carmina, Francofurti, 17‒21. Ehlers (1954): W. Ehlers, “Die Ciris und ihr Original”, MH 11, 65‒88. Erasmo (2006): M. Erasmo, “Birds of a feather? Ennius and Horace, Odes 2,20”, Latomus 65, 369‒377. Erren (2003): M. Erren, P. Vergilius Maro, Bd. 2, Heidelberg. Eskuche (1902): G. Eskuche, [Review of Skutsch (1901)], WKPh 19, 940‒946. Faber (2008): R. Faber, “The woven garment as literary metaphor: the peplos in Ciris 9‒41”, in: J. Edmondson, A. Keith (eds.), Roman Dress and the Fabrics of Roman Culture, Toronto, 205‒216. Fairclough and Goold (2000): H.R. Fairclough, G.P. Goold, Virgil: Aeneid VII–XII, Appendix Vergiliana, Cambridge, MA. Faraone (1994): C.A. Faraone, “Deianira’s mistake and the demise of Heracles: erotic magic in Sophocles’ Trachiniae”, Helios 21, 115‒135. Faraone (1999): C.A. Faraone, Ancient Greek Love Magic, Cambridge, MA. Faraone (2006): C.A. Faraone, “Magic, medicine and eros in the prologue to Theocritus’ Id. 11”, in: M. Fantuzzi, T. Papanghelis (eds.), Brill’s Companion to Greek and Latin Pastoral, Leiden, 75‒90. Farrell (1991): J. Farrell, Vergil’s Georgics and the Traditions of Ancient Epic: The Art of Allusion in Literary History, New York. Farron (1993): S. Farron, Vergil’s Aeneid: A Poem of Grief and Love, Leiden. Flores (1979): E. Flores, “La composizione dell’inno a Venere di Lucrezio e gli inni omerici ad Afrodite”, Vichiana 8, 237‒251. Foley (1991): J.M. Foley, Immanent Art: From Structure to Meaning in Traditional Oral Epic, Bloomington. Ford (1999): A. Ford, “Performing interpretation: early allegorical exegesis of Homer”, in: M. Beissinger, J. Tylus, S. Wofford (eds.), Epic Traditions in the Contemporary World: The Poetics of Community, Berkeley, 33‒53. Fowler (2002): D. Fowler, Lucretius on Atomic Motion: A Commentary on De Rerum Natura Book Two, Lines 1‒332, Oxford. Frischer (1991): B. Frischer, Shifting Paradigms: New Approaches to Horace’s Ars Poetica, Atlanta. Funke (1990): H. Funke, “Liebe als Krankheit in der griechischen und römischen Antike”, in: T. Stemmler (ed.), Liebe als Krankheit: 3. Kolloquium der Forschungsstelle für europäische Lyrik des Mittelalters, Mannheim, 11‒30. Furley (1970): D. Furley, “Variations on themes from Empedocles in Lucretius’ proem”, BICS 17, 55‒64. Galdi (1905): M. Galdi, Cornelio Gallo e la critica virgiliana, Padova. Gale (1994): M. Gale, Myth and Poetry in Lucretius, Cambridge. Gale (2001): M.R. Gale, “Etymological wordplay and poetic succession in Lucretius”, CPh 96, 158‒162. Gale (2012): M.R. Gale, “Putting on the yoke of necessity: myth, intertextuality and moral agency in Catullus 68”, in: I. Du Quesnay, T. Woodman (eds.), Catullus: Poems, Books, Readers, Cambridge, 184‒211. Gall (1999): D. Gall, Zur Technik von Anspielung und Zitat in der römischen Dichtung: Vergil, Gallus und die Ciris, München. Ganzenmüller (1984): C. Ganzenmüller, “Beiträge zur Ciris”, JKPh Suppl. 20, 551‒657.

216

Bibliography

Garner (1990): R. Garner, From Homer to Tragedy: The Art of Allusion in Greek Poetry, London. Gärtner (2008): T. Gärtner, “Medea im intertextuellen Hintergrund von Catulls Peleus-Epyllion”, AC 77, 237‒241. Gärtner (2009): U. Gärtner, “Νῆες ἀρχέκακοι: Schiffe als Unheilsbringer in der antiken Literatur”, A&A 55, 23‒44. Gatti (2008): P.L. Gatti, “La dea e la Bilancia: Elementi di datazione per la Ciris pseudovirgiliana”, CentoPagine 2, 28‒38. Gatti (2010): P.L. Gatti, Pseudo Virgilio: Ciris, Milano. Gee (2013): E. Gee, Aratus and the Astronomical Tradition, Oxford. George (1972): E.V. George, “Poet and characters in Apollonius Rhodius’ Lemnian episode”, Hermes 100, 47‒63. Gill (1973): C. Gill, “The death of Socrates”, CQ 23, 25‒28. Gill (1997): C. Gill, “Passion as madness in Roman poetry”, in: B.M. Braund, C. Gill (eds.), The Passions in Roman Thought and Literature, Cambridge, 213‒241. Gillies (1924): M. Gillies, “The ball of Eros (Ap. Rhod. III. 135 ff.)”, CR 38, 50‒51. Gorman (1995): V.B. Gorman, “Vergilian models for the characterization of Scylla in the Ciris”, Vergilius 41, 35‒48. Gow (1965): A.S.F. Gow, Theocritus, Vol. II, Cambridge. Graziosi and Haubold (2005): B. Graziosi, J. Haubold, Homer: The Resonance of Epic, London. Griffin (1981): J. Griffin, “Genre and real life in Latin poetry’, JRS 71, 39‒49. Groningen (1977): B.A. v. Groningen, Euphorion, Amsterdam. Guenzi (1994): C. Guenzi, “Il proemio della Ciris: norme retoriche, caratteri alessandrinoneoterici, tradizione romana”, Aevum(ant) 7, 301‒331. Hadjittofi (2008): F. Hadjittofi, “Callimachus’ sexy Athena: the hymn to Athena and the Homeric hymn to Aphrodite”, MD 60, 9‒37. Halliwell (1986): S. Halliwell, Aristotle’s Poetics, London. Harder (2012): A. Harder, Callimachus, Aetia: Introduction, Text, Translation, and Commentary, Vol. 2, Oxford. Hardie (1907a): W.R. Hardie, “A note on the history of the Latin hexameter”, Journal of Philology 30, 266‒279. Hardie (1907b): W.R. Hardie, “On some non-metrical arguments bearing on the date of the Ciris”, Journal of Philology 30, 280‒289. A. Hardie (2000): A. Hardie, “The ancient ‘etymology’ of ἀοιδός”, Philologus 144, 163‒175. A. Hardie (2002): A. Hardie, “The Georgics, the mysteries and the Muses at Rome”, PCPhS 48, 175‒208. A. Hardie (2009): A. Hardie, “Etymologising the Muse”, MD 62, 9‒57. A. Hardie (2013): A. Hardie, “Empedocles and the muse of the agathos logos”, AJPh 134, 209‒ 246. P. Hardie (1986): P.R. Hardie, Virgil’s Aeneid: Cosmos and Imperium, Oxford. P. Hardie (1999): P. Hardie, “Metamorphosis, metaphor, and allegory in Latin epic’, in: M. Beissinger, J. Tylus, S. Wofford (eds.), Epic Traditions in the Contemporary World: The Poetics of Community, Berkeley, 89‒107. P. Hardie (2009a): P. Hardie, Lucretian Receptions: History, the Sublime, Knowledge, Cambridge. P. Hardie (2009b): P. Hardie, “The self-divisions of Scylla”, Trends in Classics 1, 118‒147. Harrison (2007): S. Harrison, “The primal voyage and the ocean of epos: two aspects of metapoetic imagery in Catullus, Virgil and Horace”, Dictynna 4.

Bibliography

217

Heath (1988): J.R. Heath, “The blessings of epiphany in Callimachus’ Bath of Pallas”, CA 7, 72‒90. Heath (1991): M. Heath, “The universality of poetry in Aristotle’s Poetics”, CQ 41, 389‒402. Helm (1902a): R. Helm, “Vergils zehnte Ecloge”, Philologus 61, 271‒291. Helm (1902b): R. Helm, [Review of Skutsch (1901)], BPhW 22, 201‒209, 236‒242. Helm (1937): R. Helm, “Ein Epilog zur Cirisfrage”, Hermes 72, 78‒103. Henry (1950): R.M. Henry, “Pietas and fides in Catullus”, Hermathena, Vol. 75, pp. 63‒68, Vol. 76, pp. 48‒57. Heyworth (2004): S.J. Heyworth, “Looking into the river: literary history and interpretation in Callimachus, Hymns 5 and 6”, in: M.A. Harder, R.F. Reguit, G.C. Wakker (eds.), Callimachus II, Leuven, 139‒159. Hinds (1998): S. Hinds, Allusion and Intertext: Dynamics of Appropriation in Roman Poetry, Cambridge. Hollis (1982): A.S. Hollis, “Notes on Callimachus, Hecale”, CQ 32, 469‒473. Hollis (1997): A.S. Hollis, “The beginning of Callimachus’ Hecale”, ZPE 115, 55‒56. Hollis (1998): A.S. Hollis, “Nicander and Lucretius”, PLILS 10, 169‒184. Hollis (2000): A.S. Hollis, “The reputation and influence of Choerilus of Samos”, ZPE 130, 13‒ 15. Hollis (2006): A. Hollis, “Hellenistic epyllion and its descendants’, in: S.F. Johnson (ed.), Greek Literature in Late Antiquity: Dynamism, Didacticism, Classicism, Aldershot, 141‒157. Hollis (2007): A.S. Hollis, Fragments of Roman Poetry c.60 BC–AD 20, Oxford. Hollis (2009): A.S. Hollis, Callimachus: Hecale, Oxford (2nd ed.). Horsfall (2006): N. Horsfall, Virgil, Aeneid 3: A Commentary, Leiden. Höschele (2009): R. Höschele, “Catullus’ Callimachean hair-itage and the erotics of translation’, RFIC 137, 118‒152. Höschele (2011): R. Höschele, “Inscribing epigrammatists’ names: Meleager in Propertius and Philodemus in Horace”, in: A. Keith (ed.), Latin Elegy and Hellenistic Epigram: A Tale of Two Genres at Rome, Newcastle, 19‒31. Hross (1958): H. Hross, “Die Klagen der verlassenen Heroiden in der lateinischen Dichtung: Untersuchung zur Motivik und zur dichterischen Form”, Diss. München. Hughes (1929): M.Y. Hughes, Virgil and Spenser, Berkley. Hunter (1989): R.L. Hunter, Apollonius of Rhodes: Argonautica, Book III, Cambridge. Hunter (1992): R. Hunter, “Writing the god: form and meaning in Callimachus, hymn to Athena”, MD 29, 9‒34. Hunter (1993a): R. Hunter, “Callimachean echoes in Catullus 65”, ZPE 96, 179‒182. Hunter (1993b): R. Hunter, The Argonautica of Apollonius: Literary Studies, Cambridge. Hunter (1999): R. Hunter, Theocritus: A Selection, Cambridge. Hunter (2005): R. Hunter, “Generic consciousness in the Orphic Argonautica?”, in: M. Paschalis (ed.), Roman and Greek Imperial Epic, Rethymnon, 149‒168. Hutchinson (2006): G.O. Hutchinson, Propertius: Elegies, Book IV, Cambridge. Hutchinson (2007): G.O. Hutchinson, “Introduction”, in: R.O.A.M. Lyne, Collected Papers on Latin Poetry, Oxford, ix–xix. Huxley (1969): G. Huxley, “Choirilos of Samos”, GRBS 10, 12‒29. Jacobs (1837): F. Jacobs, “Fabii Planciadis Fulgentii Mythologiarum libri”, Beiträge zur ältern Litteratur 2, 416‒429. Jacques (2002): J.-M. Jacques, Nicandre, Œuvres, t. II: Les Thériaques, Paris. Jacques (2007): J.-M. Jacques, Nicandre, Œuvres, t. III: Les Alexipharmaques, Paris.

218

Bibliography

Jahn (1902): P. Jahn, “Aus Vergils Frühzeit”, Hermes 37, 161‒172. Jahn (1906): P. Jahn, “Jahresbericht über Vergil 1901‒1904 (1905)”, JAW 130, 1‒115. Jahn (1907): P. Jahn, [Review of Skutsch (1906)], BPhW 27, 37‒43. Jahn (1908): P. Jahn, “Vergil und die Ciris”, RhM 63, 79‒106. Jahn (1910): P. Jahn, “Jahresbericht über Vergil 1905‒1908 (1909)”, JAW 148, 1‒111. Jahn (1914): P. Jahn, “Jahresbericht über Vergil 1909‒1912 (1913)”, JAW 167, 357‒415. Janko (2011): R. Janko, Philodemus, On Poetry: Books 3‒4, Oxford. Jope (1985): J. Jope, “Lucretius, Cybele, and religion”, Phoenix 39, 250‒262. Kallendorf (2006): C. Kallendorf, “Allusion as reception: Virgil, Milton, and the modern reader’, in: C. Martindale, R.F. Thomas (eds.), Classics and the Uses of Reception, Malden, 67‒79. Kayachev (2012): B. Kayachev, “The so-called Orphic gold tablets in ancient poetry and poetics”, ZPE 180, 17‒37. Kayachev (2013): B. Kayachev, “The ideal biography of a Roman poet: from lusus poetici to studia philosophica”, Latomus 72, 412‒425. Kayachev (2014): B. Kayachev, “Ciris 137: an emendation”, CQ 64, 859‒861. Kayachev (2016): B. Kayachev, “Scylla the beauty and Scylla the beast: a Homeric allusion in the Ciris”, in A. Efstathiou, I. Karamanou (eds.), Homeric Receptions Across Generic and Cultural Contexts, 277‒287 (in print). Kayachev (forthcoming-a): B. Kayachev, “The sphragis of Virgil’s Georgics: constructing identity through intertextuality”, in A. Gavrielatos (ed.), Self-Presentation and Identity in the Roman World (forthcoming with Cambridge Scholars Publishing). Kayachev (forthcoming-b): B. Kayachev, “Ciris 524: an emendation”, RhM (forthcoming). Kelly (2007): A. Kelly, A Referential Commentary and Lexicon to Iliad VIII, Oxford. Kenney (1971): E.J. Kenney, Lucretius: De Rerum Natura, Book III, Cambridge. Kidd (1997): D. Kidd, Aratus: Phaenomena, Cambridge. King (1988): J.K. King, “Catullus’ Callimachean carmina, cc. 65‒116”, CW 81, 383‒392. Klooster (2012): J. Klooster, “Visualizing the impossible: the moving landscape in the Delos hymn of Callimachus”, Aitia 2. Knaack (1902): G. Knaack, “Hellenistische Studien: I. Nisos und Skylla in der hellenistischen Dichtung”, RhM 57, 205‒230. Knauer (1964): G.N. Knauer, Die Aeneis und Homer: Studien zur poetischen Technik Vergils, mit Listen der Homerzitate in der Aeneis, Göttingen. Knox (1983): P.E. Knox, “Cinna, the Ciris, and Ovid”, CPh 78, 309‒311. Knox (1990): P.E. Knox, “Scylla’s nurse”, Mnemosyne 43, 158‒159. Knox (1999): P.E. Knox, “Lucretius on the narrow road”, HSCPh 99, 275‒287. Knox (2011): P.E. Knox, “Cicero as a Hellenistic poet”, CQ 61, 192‒204. Kolář (1909): A. Kolář, “Kdo jest skladatelem básně Ciris?”, LF 36, 161‒182, 321‒334, 428‒447. Konstan (2000): D. Konstan, “A pun in Virgil’s Aeneid (4.492‒93)?”, CPh 95, 74‒76. Körte (1907): A. Körte, [Review of Skutsch (1906)], WKPh 24, 1336‒1342. Kraias (2008): G. Kraias, “Epische Szenen in tragischem Kontext: Untersuchung zu den Homer-Bezügen bei Aischylos”, Diss. Freiburg. Kreunen (1882): M. Kreunen, “Prolegomena in Cirin”, Diss. Utrecht 1882. Kubiak (1981): D.P. Kubiak, “The Orion episode of Cicero’s Aratea”, CJ 77, 12‒22. Kyriakidis (2006): S. Kyriakidis, “Lucretius’ DRN 1.926‒50 and the proem to Book 4”, CQ 56, 606‒610. Kyriakidis (2010): S. Kyriakidis, “Heroides 20 and 21: motion and emotion”, LICS 9.2, 1‒13.

Bibliography

219

Kyriakou (1994): P. Kyriakou, “Empedoclean echoes in Apollonius’ Argonautica”, Hermes 122, 309‒319. Kyriakou (1995): P. Kyriakou, Homeric Hapax Legomena in the Argonautica of Apollonius Rhodius: A Literary Study, Stuttgart. Laird (2003): A. Laird, “Figures of allegory from Homer to Latin epic”, in: G.R. Boys-Stones (ed.), Metaphor, Allegory, and the Classical Tradition: Ancient Thought and Modern Revisions, Oxford. Landolfi (1982): L. Landolfi, “Tracce filodemee di estetica e di epigrammatica simpotica in Catullo”, CErc 12, 137‒143. Lange (2003): K. Lange, Euripides und Homer: Untersuchungen zur Homer-Nachwirkung in Elektra, Iphigenie im Taurerland, Helena, Orestes, und Kyklops, Stuttgart. Lateiner (1990): D. Lateiner, “Mimetic syntax: metaphor from word order, especially in Ovid”, AJPh 111, 204‒237. Latte (1935): K. Latte, “Der Thrax des Euphorion”, Philologus 90, 129‒155. Lee (2004): M.M. Lee, “ ‘Evil wealth of raiment’: deadly πέπλοι in Greek tragedy”, CJ 99, 253‒279. Lehnus (1975): L. Lehnus, “Una scena della Ciris (vv. 220ss.): Carme e l’Ecale di Callimaco”, RIL 109, 353‒361. Leo (1902): F. Leo, “Vergil und die Ciris”, Hermes 37, 14‒55. Leo (1907): F. Leo, “Nochmals die Ciris und Vergil”, Hermes 42, 35‒77. Levin (1970): D.N. Levin, “Δίπλαξ πορφυρέη”, RFIC 98, 17‒36. Levin (1971): D.N. Levin, Apollonius’ Argonautica Re-Examined, I: The Neglected First and Second Books, Leiden. Lightfoot (2009): J.L. Lightfoot, Hellenistic Collection: Philitas, Alexander of Aetolia, Hermesianax, Euphorion, Parthenius, Cambridge, MA. Lord (1951): A.B. Lord, “Composition by theme in Homer and Southslavic epos”, TAPhA 82, 71‒80. Lord (1960): A.B. Lord, The Singer of Tales, Cambridge, MA. Louden (1999): B. Louden, The Odyssey: Structure, Narration, and Meaning, Baltimore. Luppe (2008): W. Luppe, “Textvorschläge zur pseudo-vergilianischen ‚Ciris‘ ”, Philologus 152, 161‒165. Lyne (1978): R.O.A.M. Lyne, Ciris: A Poem Attributed to Vergil, Cambridge. MacFarlane (2009): K.A. MacFarlane, “Choerilus of Samos’ lament (SH 317) and the revitalization of epic”, AJPh 130, 219‒234. Magnelli (2006): E. Magnelli, “Nicander’s chronology: a literary approach”, in: M.A. Harder, R.F. Regtuit, G.C. Wakker (eds.), Beyond the Canon, Leuven, 185‒204. Maltby (1991): R. Maltby, A Lexicon of Ancient Latin Etymologies, Leeds. Maltby (1993): R. Maltby, “Varro’s attitude to Latin derivations from Greek”, PLILS 7, 47‒60. Maltby (2005): R. Maltby, “Etymologising and the structure of argument in Lucretius Book 1”, PLLS 12, 95‒111. Maltby (2006): R. Maltby, “Major themes and motifs in Propertius’ love poetry”, in: H.-C. Günther (ed.), Brill’s Companion to Propertius, Leiden, 147‒181. Maltby (2010): R. Maltby, “The unity of Corpus Tibullianum Book 3: some stylistic and metrical considerations”, PLLS 14, 319‒340. McGill (2012): S. McGill, Plagiarism in Latin Literature, Cambridge. McKeown (1987): J.C. McKeown, Ovid: Amores, Vol. I, Liverpool.

220

Bibliography

McLennan (1977): G.R. McLennan, Callimachus, Hymn to Zeus: Introduction and Commentary, Rome. Melia (2005): D.F. Melia, “Orality and Aristotle’s aesthetics”, in: M.C. Amodio (ed.), New Directions in Oral Poetry, Tempe, 91‒105. Merriam (2001): C.U. Merriam, “Clinical cures for love in Propertius’ Elegies”, Scholia 10, 69‒ 76. Michels (1945): A.K. Michels, “The Golden Bough of Plato”, AJPh 66, 59‒63. Mignogna (1993): E. Mignogna, “Europa o Selene? Achille Tazio e Mosco, o il ritorno dell’ “inversione””, Maia 45, 177‒183. Miller (1994): J.F. Miller, “The memories of Ovid’s Pythagoras”, Mnemosyne 47, 473‒487. Minchin (2001): E. Minchin, Homer and the Resources of Memory: Some Applications of Cognitive Theory to the Iliad and the Odyssey, Oxford. Mineur (1984): W.H. Mineur, Callimachus, Hymn to Delos: Introduction and Commentary, Leiden. Mori (2005): A. Mori, “Jason’s reconciliation with Telamon: a moral exemplar in Apollonius’ Argonautica (1.1286‒1344)”, AJPh 126, 209‒236. Mori (2008): A. Mori, The Politics of Apollonius Rhodius’ Argonautica, Cambridge. Morisi (2002): L. Morisi, “Ifigenia e Polissena (Lucrezio in Catullo)”, MD 49, 177‒190. Most (2010): G.W. Most, “Hellenistic allegory and Early Imperial rhetoric”, in: R. Copeland, P.T. Struck (eds.), The Cambridge Companion to Allegory, Cambridge, 26‒38. Munari (1944): F. Munari, Studi sulla “Ciris”, Firenze. Murnaghan (1999): S. Murnaghan, “The poetics of loss in Greek epic”, in: M. Beissinger, J. Tylus, S. Wofford (eds.), Epic Traditions in the Contemporary World: The Poetics of Community, Berkeley, 203‒220. Murrin (1969): M. Murrin, The Veil of Allegory: Some Notes Toward a Theory of Allegorical Rhetoric in the English Renaissance, Chicago. Murrin (1980): M. Murrin, The Allegorical Epic: Essays in Its Rise and Decline, Chicago. Nagler (1974): M.N. Nagler, Spontaneity and Tradition : A Study in the Oral Art of Homer, Berkeley. Nagy (1990a): G. Nagy, Pindar’s Homer: The Lyric Possession of an Epic Past, Baltimore. Nagy (1990b): G. Nagy, Greek Mythology and Poetics, Ithaca. Nagy (2003): G. Nagy, Homeric Responses, Austin. Nagy (2010): G. Nagy, Homer the Preclassic, Berkeley. Nelis (1992): D.P. Nelis, “Demodocus and the song of Orpheus: Ap. Rhod. 1, 496‒511”, MH 49, 153‒170. Nelis (2001): D. Nelis, Vergil’s Aeneid and the Argonautica of Apollonius Rhodius, Leeds. Némethy (1907): G. Némethy, “Zur Ciris-Frage”, RhM 62, 482‒485. Nishimura-Jensen (1998): J. Nishimura-Jensen, “The poetics of Aethalides: silence and poikilia in Apollonius’ Argonautica”, CQ 48, 456‒469. Nishimura-Jensen (2000): J. Nishimura-Jensen, “Unstable geographies: the moving landscape in Apollonius’ Argonautica and Callimachus’ Hymn to Delos”, TAPhA 130, 287‒317. Norden (1926): E. Norden, P. Vergilius Maro: Aeneis Buch VI, Leipzig (3rd ed.). Nünlist (1998): R. Nünlist, Poetologische Bildersprache in der frühgriechischen Dichtung, Stuttgart. Nyberg (1992): L. Nyberg, Unity and Coherence: Studies in Apollonius Rhodius’ Argonautica and the Alexandrian Epic Tradition, Lund.

Bibliography

221

O’Hara (1996): J.J. O’Hara, True Names: Vergil and the Alexandrian Tradition of Etymological Wordplay, Ann Arbor. O’Hara (2007): J.J. O’Hara, Inconsistency in Roman Epic: Studies in Catullus, Lucretius, Vergil, Ovid and Lucan, Cambridge. O’Sullivan (1981): J.N. O’Sullivan, “Asius and the Samians’ hairstyle”, GRBS 22, 329‒333. Obbink (2010): D. Obbink, “Early Greek allegory”, in: R. Copeland, P.T. Struck (eds.), The Cambridge Companion to Allegory, Cambridge, 15‒25. Oliensis (1998): E. Oliensis, Horace and the Rhetoric of Authority, Cambridge. Overduin (2015): F. Overduin, Nicander of Colophon’s Theriaca: A Literary Commentary, Leiden. Panussi (2009): V. Panussi, Greek Tragedy in Vergil’s “Aeneid”: Ritual, Empire, and Intertext, Cambridge. Papanghelis (1994): T. Papanghelis, “Hoary ladies: Catullus 305 ff. and Apollonius of Rhodes”, SO 69, 41‒46. Parlato (2008): G. Parlato, “I fiori di Afrodite: nota a Cypr. fr. 4, 6 Bernabé”, QUCC 88, 145‒ 149. Peirano (2012): I. Peirano, The Rhetoric of the Roman Fake: Latin Pseudepigrapha in Context, Cambridge. Pendergraft (1991): M.L.B. Pendergraft, “Eros ludens: Apollonius’ Argonautica 3, 132‒41”, MD 26, 95‒102. Perotti (2006): P.A. Perotti, “Note al c. 76 di Catullo”, GIF 58, 293‒311. Pigeaud (1983): J. Pigeaud, “La métamorphose de Scylla (Ciris 490‒507)”, LEC 51, 125‒131. Platt (1920): A. Platt, “Apollonius III”, The Journal of Philology 35, 72‒85. Pontani (1999): F. Pontani, “The first word of Callimachus’ Aitia”, ZPE 128, 57‒59. Pontani (2011): F. Pontani, “”El universo es, como tú, Proteo”: selected readings of a Homeric myth”, A&A 57, 129‒150. Pucci (1987): P. Pucci, Odysseus Polutropos: Intertextual Readings in the Odyssey and the Iliad, Ithaca. Ross (1975): D.O. Ross, Backgrounds to Augustan Poetry: Gallus, Elegy and Rome, Cambridge. Ross (1987): D.O. Ross, Virgil’s Elements: Physics and Poetry in the Georgics, Princeton. Ruiz Sánchez (1997): M. Ruiz Sánchez, “Formal technique and epithalamial setting in the song of the Parcae (Catullus 64.305‒22, 328‒36, 372‒80)”, AJPh 118, 75‒88. Rumpf (2005): L. Rumpf, “Lukrez und Parmenides”, Philologus 149, 78‒95. Russell (1979): D.A. Russell, “De imitatione”, in: D. West, T. Woodman (eds.), Creative Imitation and Latin Literature, Cambridge, 1‒16. Rynearson (2009): N. Rynearson, “A Callimachean case of lovesickness: magic, disease, and desire in Aetia frr. 67‒75 Pf.”, AJPh 130, 341‒365. Salvatore (1984): A. Salvatore, “Echi degli Aratea nella Ciris”, Ciceroniana 5, 237‒241. Sayers (1941): D.L. Sayers, The Mind of the Maker, London. Scaliger (1573): I. Scaliger, Publii Virgilii Maronis appendix, Lugduni. Scheid and Svenbro (1996): J. Scheid, J. Svenbro, The Craft of Zeus: Myths of Weaving and Fabric, Cambridge, MA. Schiesaro (2008): A. Schiesaro, “Furthest voices in Virgil’s Dido”, SIFC 6, 60‒109, 194‒245. Schiesaro (2009): A. Schiesaro, “Horace’s Bacchic poetics”, in: L.B.T. Houghton, M. Wyke (eds.), Perceptions of Horace: A Roman Poet and His Readers, Cambridge, 61‒79. Schmakeit (2003): I.A. Schmakeit, “Apollonios Rhodios und die attische Tragödie: Gattungsüberschreitende Intertextualität in der alexandrinischen Epik”, Diss. Groningen.

222

Bibliography

Schmiel (1981): R. Schmiel, “Moschus’ Europa”, CPh 4, 261‒272. Schwabe (1871): L. Schwabe, In Cirin carmen observationum pars I, Dorpat. Seeliger (1893): K. Seeliger, “Alkathoos und die megarische Königliste: Eine Pausaniasstudie”, in: Festschrift für Johannes Overbeck, Leipzig, 27‒44. Setaioli (1975): A. Setaioli, “Un influsso ciceroniano in Virgilio”, SIFC 47, 5‒26. Shapiro (1980): H.A. Shapiro, “Jason’s cloak”, TAPhA 110, 263‒286. Sider (1987): D. Sider, “The love poetry of Philodemus”, AJPh 108, 310‒324. Sider (1995): D. Sider, “The Epicurean philosopher as Hellenistic poet”, in: D. Obbink (ed.), Philodemus and Poetry: Poetic Theory and Practice in Lucretius, Philodemus, and Horace, New York, 42‒57. Sider (1997): D. Sider, The Epigrams of Philodemos: Introduction, Text, and Commentary, New York. Skinner (1976): M.B. Skinner, “Iphigenia and Polyxena: a Lucretian allusion in Catullus”, Pacific Coast Philology 11, 52‒61. Skinner (2003): M.B. Skinner, Catullus in Verona: A Reading of the Elegiac Libellus, Poems 65‒116, Columbus. Skinner (2005): M.B. Skinner, “Homer’s mother”, in: E. Greene (ed.), Women Poets in Ancient Greece and Rome, Norma, 91‒111. Skutsch (1901): F. Skutsch, Aus Vergils Frühzeit, Leipzig. Skutsch (1906): F. Skutsch, Gallus und Vergil, Leipzig. Slings (2004): S.R. Slings, “The Hymn to Delos as a partial allegory of Callimachus’ poetry”, in: M.A. Harder, R.F. Regtuit, G.C. Wakker (eds.), Callimachus II, Leuven, 279‒298. Smart (2012): J. Smart, “Intertextual dynamics in Moschus’s Europa”, Arethusa 45, 43‒55. Solodow (1988): J.B. Solodow, The World of Ovid’s Metamorphoses, Chapel Hill. Sonntag (1902): H. Sonntag, “Virgil und Cornelius Gallus”, WKPh 19, 1068‒1078. Spanoudakis (2002): K. Spanoudakis, Philitas of Cos, Leiden. Spanoudakis (2004): K. Spanoudakis, “Adesp. Pap. Eleg. SH 964: Parthenius?”, APF 50, 37‒41. Stephens (2015): S.A. Stephens, Callimachus: The Hymns, Oxford. Stückelberger (1990): A. Stückelberger, “Sterngloben und Sternkarten: Zur wissenschaftlichen Bedeutung des Leidener Aratus”, MH 47, 70‒81. Sudhaus (1907): S. Sudhaus, “Die Ciris und das römische Epyllion”, Hermes 42, 469‒504. Tait (1941): J.I.M. Tait, Philodemus’ Influence on the Latin Poets, Bryn Mawr, PA. Tarrant (1981): R.J. Tarrant, “The authenticity of the letter of Sappho to Phaon (Heroides XV)”, HSCPh 85, 133‒153. Thévenaz (2002): O. Thévenaz, “Le cygne de Venouse: Horace et la métamorphose de l’Ode II, 20”, Latomus 61, 861‒888. van Thiel (1996): H. van Thiel, Homeri Ilias, Hildesheim. Thill (1979): A. Thill, Alter ab illo: Recherches sur l’imitation dans la poésie personelle à l’époque Augustéenne, Paris. Thomas (1979): R.F. Thomas, [Review of Lyne (1978)], Phoenix 33, 180‒184. Thomas (1981): R.F. Thomas, “Cinna, Calvus, and the Ciris”, CQ 31, 371‒374. Thomas (1982a): R.F. Thomas, “Catullus and the polemics of poetic reference (poem 64.1‒ 18)”, AJPh 103, 144‒164. Thomas (1982b): R.F. Thomas, “Gadflies (Virg. Geo. 3.146‒148)”, HSCPh 86, 81‒85. Thomas (1983): R.F. Thomas, “Virgil’s ecphrastic centerpieces”, HSCPh 87, 175‒184. Thomas (1986): R.F. Thomas, “Virgil’s Georgics and the art of reference”, HSCPh 90, 171‒198.

Bibliography

223

Thomas (1993): R.F. Thomas, “Callimachus back in Rome”, in: M.A. Harder, R.F. Regtuit, G.C. Wakker (eds.), Callimachus, Groningen, 197‒215. Thornton (1984): A. Thornton, Homer’s Iliad: Its Composition and the Motif of Supplication, Göttingen. Tilg (2006): S. Tilg, “Cantores Euphorionis – in Epikurs Garten?”, Philologus 150, 61‒84. Toohey (1990): P. Toohey, “Some ancient histories of literary melancholia’, ICS 15, 143‒161. Toohey (1992): P. Toohey, “Love, lovesickness, and melancholia’, ICS 17, 265‒286. Toohey (2004): P. Toohey, Melancholy, Love, and Time: Boundaries of the Self in Ancient Literature, Ann Arbor. Tränkle (1990): H. Tränkle, Appendix Tibulliana, Berlin. Trappes-Lomax (2007): J.M. Trappes-Lomax, Catullus: A Textual Reappraisal, Swansea. Trépanier (2003): S. Trépanier, Empedocles: An Interpretation, New York. Trimble (2010): G. Trimble, “A commentary on Catullus 64, lines 1‒201”, Diss. Oxford. Tsagalis (2004): C. Tsagalis, Epic Grief: Personal Laments in Homer’s Iliad, Berlin. Tschiedel (2002): H.J. Tschiedel, [Review of Gall (1999)], AAHG 55, 177‒180. Usher (1998): M.D. Usher, Homeric Stitchings: The Homeric Centos of the Empress Eudocia, Lanham. Vian (1974): F. Vian, Apollonios de Rhodes, Argonautiques, t. I: Chants I–II, Paris. Vollmer (1906): F. Vollmer, “Zu Vergils 6. Ekloge”, RhM 61, 481‒490. Voss (1797): J.H. Voss, Des Publius Vergilius Maro Ländliche Gedichte, 2 Bd., Altona. Waltz (1881): A. Waltz, De carmine Ciris, Paris. Waszink (1947): J.H. Waszink, “Varia critica et exegetica”, Mnemosyne 13, 121‒129. Watkins (1995): C. Watkins, How to Kill a Dragon: Aspects of Indo-European Poetics, Oxford. Weber (1995): C. Weber, “The allegory of the Golden Bough”, Vergilius 41, 3‒34. West (1983): M.L. West, The Orphic Poems, Oxford. West (1987): D. West, The Bough and the Gate, Exeter. West (1988): S. West, “Books I–IV”, in: A. Heubeck, S. West, J.B. Hainsworth, A Commentary on Homer’s Odyssey, vol. I: Introduction and Books I–VIII, Oxford, 51‒245. West (2007): M.L. West, Indo-European Poetry and Myth, Oxford. Whalley (1997): G. Whalley, Aristotle’s Poetics, Montreal. White (1979): H. White, Theocritus’ Idyll XXIV: A Commentary, Amsterdam. Wiesen (1981): D.S. Wiesen, “Two problems in Horace’s Satires”, Mnemosyne 34, 87‒95. Wills (1996): J. Wills, Repetition in Latin Poetry: Figures of Allusion, Oxford. Wimmel (1960): W. Wimmel, Kallimachos in Rom: Die Nachfolge seines apologetischen Dichtens in der Augusteerzeit, Wiesbaden. Woodman (2012): T. Woodman, “A covering letter: poem 65”, in: I. Du Quesnay, T. Woodman (eds.), Catullus: Poems, Books, Readers, Cambridge, 130‒152. Woytek (2005): E. Woytek, “Anmerkungen zur Catull-Rezeption in der Ciris”, in: I. Tar, P. Mayer (eds.), Studia Catulliana: In memoriam Stephani Caroli Horvath (1931‒1966), Szeged, 77‒89. Wright (1981): M.R. Wright, Empedocles: The Extant Fragments, New Haven.

Index of passages cited Achilles Tatius – 1.4.4: 164 Aeschylus, Agamemnon – 228: 133 Anthologia Graeca – AP 5.112: see Philodemus – AP 5.120: see Philodemus – AP 11.41: see Philodemus – AP 11.218.2: 41 – APl 279: 57 – 58 Apollonius, Argonautica – 1.1: 55 – 56 – 1.4: 56 – 1.5: 61 – 1.219 – 223: 177 – 1.278 – 291: 122 – 123 – 1.499: 189 – 1.508: 78 – 1.509 – 510: 43 – 1.513 – 514: 149 – 1.536 – 541: 148 – 1.547 – 552: 148 – 1.573 – 579: 148 – 149 – 1.611 – 613: 116 – 1.621 – 622: 189 – 1.640 – 696: 118 n. 59 – 1.640 – 649: 117 – 118, 187 – 189 – 1.641 – 642: 190 n. 30 – 1.644 – 645: 190 n. 30 – 1.647 – 648: 73, 189 n. 26 – 1.647: 189 – 1.721 – 724: 42 – 44 – 1.725 – 726: 43, 45 n. 74 – 1.730 – 731: 43 – 1.768: 42 – 1.774: 44 – 1.813 – 819: 116 – 1.804 – 806: 116 – 1.1261 – 1265: 81 – 1.1262: 85 – 1.1269: 81

– 1.1317 – 1318: 121 n. 68 – 2.197 – 201: 62 – 2.273 – 283: 177 – 2.1069 – 1072: 185 – 187 – 2.1083 – 1084: 197 – 2.1083: 185 – 187 – 2.1088 – 1089: 185 – 187 – 2.1088: 197 – 2.1247 – 1259: 202 – 203 – 2.1277 – 1278: 68 – 3.92 – 94: 70 – 3.97: 70 – 3.129 – 144: 70 – 3.132 – 133: 74 – 3.132: 78 – 3.134: 78 – 3.138: 71 – 3.152: 78 – 3.275 – 287: 70 – 3.275 – 277: 80 – 3.279: 80 – 3.286 – 287: 80 – 3.297 – 298: 80 – 3.471: 164 – 3.616 – 824: 94 – 3.616 – 651: 109 – 3.645: 104, 110 – 3.646: 96, 105 – 3.649: 81 n. 67 – 3.662: 110 – 3.664 – 665: 110 – 3.670 – 673: 110 – 3.674 – 676: 84 – 3.681 – 682: 110 – 3.688: 110 – 3.695 – 696: 110 – 3.697 – 704: 111 – 3.716 – 717: 111 – 3.734: 94 – 3.744 – 751: 94, 109 – 3.751: 100, 103 – 3.755: 104, 111 – 3.760: 104, 111 – 3.766 – 767: 111

Index of passages cited

– 3.775 – 777: 112 – 3.790: 111 n. 51 – 3.806 – 807: 111 n. 51 – 3.839: 109 – 3.867: 5, 112 – 3.1013 – 1014: 112 – 3.1013: 5 – 3.1029: 96 – 3.1193 – 1197: 95 – 4.43: 96, 105 – 4.48 – 49: 95 – 4.156 – 159: 141 – 4.355 – 390: 161 – 162 – 4.358 – 359: 162 – 4.360 – 363: 161 – 4.362 – 363: 162 – 4.368: 162 – 4.373 – 374: 162 – 4.376: 162 – 4.380: 161 – 4.386 – 387: 162 – 4.421 – 422: 139 – 4.428 – 429: 139 – 4.445 – 449: 78 – 4.445: 167 – 4.460: 139 – 4.468 – 469: 139 – 4.478 – 479: 139 – 4.770 – 771: 195 n. 39 – 4.827 – 829: 64 – 4.827 – 828: 177 – 4.481: 139 – 4.825 – 832: 150 – 4.838 – 841: 146 – 4.913 – 915: 150 n. 20 – 4.917 – 919: 150 n. 20 – 4.917 – 918: 196 – 4.924 – 929: 72 – 73 – 4.930: 148 – 4.940: 70 n. 48 – 4.948 – 952: 70 – 4.950: 71 – 4.951: 73 – 4.952: 74 – 4.956 – 960: 150 – 4.1031 – 1052: 161 – 162 – 4.1036 – 1037: 161

– 4.1036: 162 – 4.1042: 162 – 4.1045 – 1046: 161 – 4.1047: 162 – 4.1537 – 1541: 174 – 175 – 4.1546 – 1547: 174 Aratus, Phaenomena – 313: 204 n. 67 – 315: 204 n. 67 Aristaenetus, Letters – 1.10.18 – 21: 83 – 1.10.28: 82 – 1.10.44 – 47: 82 Aristophanes, Frogs – 125 – 126: 192 n. 33 Aristotle On Generation and Corruption – 315b: 47 Poetics – 1451b: 97, 191, 206 – 207 Rhetoric – 1415a: 41 [Aristotle], Problems – 953a–b: 90 – 91 Asius of Samos (Bernabé) – fr. 13.1 – 6: 176 Bion (Reed) – fr. 14: 167 – 168 Callimachus Aetia (Pfeiffer) – fr. 1: 20 – fr. 1.1: 29 – fr. 1.1 – 2: 22 – fr. 1.16: 23 – fr. 1.23 – 24: 23 – fr. 1.27 – 28: 37 – fr. 4: 60 n. 21 – fr. 67.1 – 3: 83 – fr. 67.6: 70 n. 46

225

226

Index of passages cited

– fr. 110.47: 112 Hecale (Hollis) – fr. 1 – 2: 59 – 60 – fr. 17.3 – 4: 121 – fr. 49.2 – 3: 121 – 122 – fr. 49.5: 122 – fr. 49.7: 122 – fr. 49.10 – 13: 122 – 123 – fr. 90.2: 186 n. 18 – fr. 113: 93 – fr. 122: 124 Hymns – 1.60: 50 – 1.70 – 75: 169 – 170 – 1.74 – 75: 6 – 1.75: 170 n. 71 – 1.76 – 79: 171 – 3.29 – 31: 165 – 3.30 – 31: 67 n. 35 – 3.189 – 191: 126 – 127 – 3.194: 127 – 3.195 – 199: 126 – 128 – 4.11 – 18: 152 – 153 – 4.14: 5 – 4.28 – 29: 49 – 51, 151 – 4.30 – 36: 155 – 4.36 – 38: 127 – 4.41 – 50: 154 – 155 – 4.103 – 105: 157 – 4.137 – 140: 157 – 4.156: 155 n. 36 – 4.158 – 159: 155 – 4.196: 156 – 4.198: 155 – 4.242 – 243: 154 – 4.244 – 248: 127 – 4.268 – 271: 155 – 4.300 – 301: 152 – 5.57 – 136: 23 – 5.82 – 84: 23 – 5.89 – 90: 23 – 5.94: 23 Catullus – 16.5 – 6: 28 – 63.68: 87 n. 88 – 64.5: 67

– 64.8: 44 – 64.9 – 10: 43 – 64.9: 145 – 64.14 – 15: 148 – 64.19 – 21: 148 – 64.31 – 37: 156 – 157 – 64.83: 64 n. 29 – 64.85 – 86: 64 n. 29, 65 – 64.86: 80 – 64.91 – 93: 65 – 64.92 – 93: 80 – 64.94 – 98: 70 – 64.94 – 95: 167 – 64.103 – 104: 99 – 64.113 – 114: 99 – 64.120: 100 n. 21 – 64.122: 100 – 64.123: 100 n. 20 – 64.132 – 201: 159 – 64.153: 161 – 64.160 – 161: 160 – 64.164: 160 – 64.175 – 176: 160, 168 – 64.188 – 191: 160 – 64.200 – 201: 159 n. 50 – 64.215 – 224: 120 – 121 – 64.251 – 264: 78 – 64.278 – 279: 157 – 64.285 – 287: 157 – 64.297: 203 n. 60 – 64.306 – 309: 62 – 64.307 – 310: 175 – 176 – 64.309: 186 n. 18 – 64.310: 62 n. 24 – 64.314: 62 n. 24 – 64.315: 62 n. 24, n. 25 – 64.319: 62 n. 24 – 64.322 – 381: 128 – 130 – 64.328 – 329: 128 – 64.355: 129 – 64.360: 129 – 64.367 – 370: 129 – 64.372: 130 – 64.376 – 380: 129 – 65.1 – 4: 21 – 23 – 65.1 – 2: 29 – 65.8 – 11: 23

Index of passages cited

– 65.12: 25, 26 – 65.13 – 14: 23 – 65.15 – 16: 21, 22 – 65.17 – 18: 29 – 65.19 – 20: 69 – 65.19: 82 – 65.21: 82 n. 68 – 65.23: 82 n. 68 – 65.24: 81 – 82 – 65.53: 204 n. 68 – 66.1: 22 – 66.45: 113 – 66.47: 112 – 68.1 – 40: 25 n. 20 – 68.10: 27 – 68.11 – 14: 25 – 68.15 – 26: 32 – 68.15 – 20: 26 – 27 – 68.17 – 20: 113 – 68.17: 26 n. 21 – 68.25 – 26: 27 n. 23 – 68.51: 26 n. 21, 113 n. 55 – 68.96: 32 – 68.138 – 141: 66 – 67 – 68.145 – 146: 26 n. 22 – 76.26: 131 – 101.7 – 10: 25 – 116.1 – 2: 22 Choerilus of Samos – SH 316: 41 – SH 317: 40 – 41 Cicero Aratea (Soubiran) – fr. 31: 130 n. 8 – fr. 31.2: 130 – fr. 34.48: 200, 204 – fr. 34.86 – 87: 204 n. 67 – fr. 34.88: 204 – fr. 34.420: 205 – fr. 34.431 – 432: 205 De consulatu suo (Courtney) – fr. 10.71 – 78: 31 De fato – 34 – 35: 69

De inventione – 1.91: 69 n. 44 [Cicero], Rhetoric to Herennius – 1.1: 29 – 2.34: 68 – 69 Ciris – 1 – 100: 18 n. 65, 19, 20 – 54 – 1 – 53: 20, 38 – 1 – 11: 21, 38 – 1 – 2: 30, 35 – 1: 33, 113 – 2 – 3: 28 – 4: 31, 35 – 5: 24, 33 – 6: 28, 113 – 7 – 8: 114 – 7: 22, 205 – 8: 33, 35, 37 – 9: 28, 38 n. 55, 46, 199 – 10 – 11: 29 – 11: 33, 37, 54 – 12 – 41: 38 – 14 – 45: 38 – 14 – 17: 33 – 14: 35, 37, 44 – 15: 31 – 16: 36, 47 n. 81 – 18 – 21: 49 – 19 – 20: 29 n. 29 – 19: 28, 113 – 20: 151 n. 22 – 21 – 41: 145 – 21 – 26: 42 – 43 – 21: 46 – 25 – 26: 145 n. 6 – 27 – 28: 40 – 29: 46 – 32: 43 n. 67 – 35: 42 – 36 – 39: 46 – 49 – 37 – 38: 205 – 37: 43 – 39: 47, 151 n. 22, 199 – 40 – 41: 47 n. 80 – 42 – 53: 38

227

228

Index of passages cited

– 42 – 43: 54 – 42 – 50: 39 – 42 – 46: 25 – 42: 41 – 47: 38 n. 56 – 48 – 51: 173 – 51: 159 n. 45 – 52: 82 n. 72 – 54 – 100: 20 – 54 – 61: 48 – 54 – 57: 48 – 50, 151 – 59 – 61: 49 – 62 – 91: 48 – 70 – 76: 178 – 89 – 90: 50 – 90 – 91: 150, 178 – 92 – 100: 41, 51 – 92: 54 – 94 – 95: 132 – 99 – 100: 47 n. 80 – 100: 47 – 101 – 205: 18 n. 65, 19, 55 – 91 – 101 – 109: 56 – 57 – 101 – 105: 59 – 60 – 101 – 103: 47 – 48 – 101: 56 – 102 – 103: 205 – 105 – 109: 57 – 59 – 110 – 115: 56, 60 – 110 – 111: 58 – 116 – 128: 56 – 119 – 125: 61 – 120 – 122: 175 – 177 – 122: 186 n. 18 – 126 – 128: 175 – 126 – 129: 62 n. 24 – 127 – 128: 62 n. 25 – 129 – 132: 18 n. 65, 57, 63 – 64 – 130 – 131: 177 – 132: 65 – 133 – 157: 66 – 133 – 134: 70 – 137 – 139: 66 – 67 – 137: 66 – 67 n. 34 – 138 – 139: 205 – 206 – 139: 79 – 141: 79

– 142 – 143: 71, 76, 138 – 144 – 145: 78 – 79 – 144: 71 – 145: 176 – 149 – 155: 67 – 149 – 152: 78 – 149 – 150: 69 – 150: 77 – 154 – 155: 79 – 154: 205 – 158 – 190: 79 – 158 – 162: 70 – 160: 80 – 161 – 162: 83 – 161: 90 – 162 – 163: 80 – 163 – 164: 66, 80, 86, 205 – 165 – 168: 87 – 169: 91 n. 103 – 171: 86 – 172: 114 – 175: 168 – 180 – 184: 80 – 81 – 180: 82 n. 68 – 181 – 182: 91, 137 – 182: 85, 86 – 184: 82 n. 68 – 185 – 186: 137 – 186: 82 – 187: 82 n. 68 – 189: 82 n. 72 – 191 – 205: 18 n. 65 – 191: 158 – 198 – 202: 158 – 201: 95 – 206 – 348: 18 n. 65, 19, 92 – 127 – 206 – 219: 93 – 94, 110 – 206 – 209: 103 – 206 – 207: 92 – 206: 100 – 209: 100 n. 20, 104, 109 – 210: 105 – 211: 104, 110 – 212: 96, 109 – 213 – 214: 99 – 213: 105 – 214: 81 n. 67, 109

Index of passages cited

– 216 – 217: 109 – 217 – 218: 114 – 218: 96 – 219: 99, 105 – 220 – 221: 99 – 221 – 223: 110 – 221: 106 – 223: 106 – 224 – 249: 114 – 225 – 226: 83 – 85 – 229 – 231: 84 n. 78, n. 80 – 231 – 236: 106 – 234 – 235: 115 – 235: 130 – 237 – 238: 83 – 84 – 237: 91 – 239: 82 n. 72 – 241 – 243: 113 – 242 – 243: 26 n. 21, 84, 167 – 244: 115 – 252: 105 n. 35, 109 – 253: 110 – 255 – 256: 110 – 256: 107 – 257 – 282: 114 – 258: 107 – 259 – 264: 115 – 116, 167 – 262: 110 – 264: 107 – 266: 107 – 268 – 275: 107 – 268 – 272: 116 – 118 – 273 – 277: 111 – 278 – 282: 112 – 280: 5, 107 – 282: 107 – 283: 110 – 284 – 289: 119 – 121 – 285: 114, 157 – 286 – 339: 114, 119 – 291: 121 – 292: 120 – 294 – 296: 122 – 123 – 301 – 305: 126 – 310 – 314: 124 – 125 – 311: 122 – 123 – 314: 123

– 316 – 317: 121 – 318: 123 – 323: 82 n. 72 – 326: 123 – 338 – 339: 111 – 345: 104, 111 – 349 – 385: 18 n. 65, 19, 128 – 144 – 349 – 352: 92 – 349: 129 – 353: 129 – 354: 130 – 355: 133 – 360 – 361: 129 – 130 – 362 – 363: 139 – 363: 131 – 365 – 368: 129 – 366: 139, 144 – 368: 131 – 369: 142 – 371 – 373: 139 – 376 – 377: 141 – 378: 137 – 380: 61, 142 n. 35 – 381: 147 – 384 – 385: 146 n. 9 – 386 – 477: 18 n. 65, 19, 145 – 172 – 386 – 390: 142 – 144 – 386: 147 – 387 – 389: 148 – 389: 149, 163 – 391 – 399: 149 – 391 – 392: 147 – 391: 151 – 392: 152 – 400: 115 – 401: 114 – 115, 157 – 404 – 458: 114, 145 – 404 – 415: 159 – 405 – 407: 160 – 408 – 411: 158 – 159 – 414 – 427: 167 – 414 – 415: 161 – 416 – 437: 159 – 416: 162 – 418 – 420: 161 – 419 – 420: 162 – 421: 82 n. 72

229

230

Index of passages cited

– 423 – 424: 158 – 424: 161 – 427: 161 – 428 – 437: 167 – 428 – 432: 163 – 428: 162 – 429: 168 – 430: 6, 164 – 431 – 432: 160, 162 n. 54, 167 – 168, 506 n. 72 – 433 – 437: 169 – 170 – 437: 6, 170 n. 71, 191 – 438 – 458: 167 – 438 – 447: 159 – 440: 158 n. 43 – 441 – 442: 161 – 441: 167 – 444 – 446: 160 – 445 – 447: 162 – 448 – 458: 159 – 448 – 450: 160 – 449 – 450: 182 – 454 – 458: 160 – 458: 161 – 462: 115 – 463 – 477: 154 – 155 – 472: 82 n. 72 – 473 – 476: 156 n. 38 – 478 – 541: 18 n. 65, 19, 173 – 208 – 478 – 489: 173 – 478 – 480: 174 – 175 – 481: 175, 196 – 482 – 483: 150 n. 20 – 483: 179 – 484 – 488: 150, 177 – 178 – 487 – 489: 126, 195 – 196 – 487: 196 – 490 – 507: 173 – 490 – 495: 180 – 490 – 492: 181 n. 14 – 491 – 492: 201 n. 57 – 491: 183 – 494 – 495: 201 n. 57 – 496 – 498: 181 – 499: 182 – 500 – 504: 183 – 186 – 502: 197, 201

– 503 – 597: 182 – 504: 197 – 506: 198 – 508 – 509: 179 – 510 – 541: 173 – 510: 206 n. 71 – 512: 158 n. 43, 179 – 516: 5 – 514 – 519: 153 – 514 – 516: 195 – 196 – 515 – 516: 202 – 519: 203 – 520 – 527: 131 – 133 – 524 – 529: 204 – 527: 182, 207 n. 73 – 532 – 537: 205 – 206 – 533 – 537: 203 – 534: 6 n. 25 – 536 – 537: 204 n. 65 – 538 – 541: 4, 195, 206 – 539: 203 – 541: 199 Clement, Protrepticus – 2.17: 77 – 78, 138 – 139 Cypria (Bernabé) – fr. 4: 52 – 53 Empedocles (Diels – Kranz) – fr. 3.6 – 13: 34 n. 41 – fr. 3.6 – 8: 34 – 35 – fr. 16.2: 199 n. 49 – fr. 17.11: 189 n. 25 – fr. 26.10: 189 n. 25 – fr. 27.3: 71 – fr. 112: 136 – fr. 115: 192 – fr. 115.3 – 4: 79, 206 – fr. 115.6 – 13: 188 – 189 – fr. 115.6 – 7: 199 n. 49 – fr. 115.6: 206 n. 71 – fr. 115.9 – 12: 73 – fr. 115.9 – 10: 206 – fr. 115.11: 190 n. 30 – fr. 115.12: 206 – fr. 124: 35

Index of passages cited

– fr. 128.2 – 3: 132 – fr. 128.8 – 10: 132 – fr. 129: 136 – fr. 129.4: 36 n. 49 – fr. 137: 132 – 133 Ennius Annals (Skutsch) – fr. 1.8: 183 – fr. 1.9: 183 Medea Exul (Jocelyn) – fr. 103: 67 – 68

Eustathius, Commentarii ad Homeri Odysseam – 1.241 – 242: 73 – 74 Fulgentius, Mythologies – 2.9: 82 n. 72 Germanicus, Aratus – 647: 50 n. 90 Heraclitus, Homeric Problems – 70.11: 82 n. 72

Etymologicum Gudianum – s.v. ἀετός: 204 n. 65

Herodotus, Histories – 1.2: 68

Etymologicum magnum – s.v. ἀετός: 203 n. 61

Hesiod Theogony – 27 – 28: 50 – 775 – 806: 190 n. 30 Works and Days – 465: 204 n. 64

Euphorion (Lightfoot) Hyacinthus – fr. 44: 49 Thrax – fr. 26.i.10: 110 – fr. 26.ii.14 – 19: 63 – 64 Euripides Hecuba – 472 – 474: 43 Hippolytus – 136 – 138: 84 n. 80 – 141 – 144: 87 – 347 – 352: 84 – 523 – 524: 111 n. 49 Medea – 1 – 8: 67 – 68 – 465 – 519: 161 – 162 – 483 – 485: 162 – 502 – 503: 162 – 518 – 519: 162 n. 54 – 776: 163 – 869 – 905: 163 – 956 – 958: 162 – 1343: 162

Hippocrates, De natura pueri – 13, 17, 19: 181 Homer Iliad – 1.1: 41 – 2.1 – 4: 108 n. 42 – 3.125 – 131: 45 – 46 – 3.126: 45 n. 74 – 4.426: 140 n. 30, 153 – 6.288 – 296: 44 – 6.407 – 439: 125 n. 80 – 8.480: 73 – 9.473: 109 – 10.1 – 4: 103, 108 n. 42 – 10.1 – 2: 106 – 10.2: 102 – 10.9 – 10: 104 – 10.12 – 13: 108 n. 42 – 10.21: 104 – 10.22: 105 n. 36 – 10.26: 102, 103 – 10.82 – 85: 106 – 10.91 – 92: 103

231

232

Index of passages cited

– 10.94 – 95: 104 – 10.97 – 99: 104 – 10.132: 105 n. 36 – 10.141 – 142: 106 – 10.181 – 182: 104 – 10.251 – 253: 102, 108 n. 42 – 10.252: 105 – 10.256 – 258:105 – 10.261 – 262: 105 – 10.275 – 295: 105 – 10.308 – 312: 104 n. 32 – 10.339: 106 – 10.376 – 377: 106 – 10.378: 107 – 10.384 – 389: 106 – 10.390 – 391: 107 – 10.395 – 399: 104 n. 32 – 10.408: 104 – 10.413: 107 – 10.416 – 421: 104 – 10.417: 108 n. 42 – 10.427: 107 – 10.432: 107 – 10.435 – 441: 107 – 10.437: 108 n. 42 – 10.439: 108 n. 42 – 10.447: 107 – 10.454 – 456: 107 – 10.474: 104, 107 – 10.477: 107 – 108 – 10.496 – 497: 108 n. 42 – 13.685: 176 n. 6 – 14.294: 164 – 14.315 – 328: 165 – 16.188: 73 – 16.391: 186 n. 18 – 17.79 – 81: 189 – 17.371 – 372: 73 – 17.547: 186 n. 18 – 22.134: 73 – 22.437 – 438: 125 – 22.438: 124 – 22.441: 45 n. 74 – 22.451: 125 – 22.454: 124 – 125 – 23.697: 140 – 23.781: 140 n. 30

– 24.343: 95 – 24.362 – 363: 106 – 24.443 – 445: 95 – 24.677 – 681: 95 – 24.721: 152 Odyssey – 1.1: 41, 55 – 1.261 – 262: 88 – 2.181: 73 – 2.228 – 230: 88 – 4.406: 5, 153 n. 30 – 5.322 – 323: 140 – 5.334: 140 – 6.95: 140 n. 30 – 6.98: 73 – 6.102 – 109: 74 – 6.115 – 117: 71 – 6.123 – 124: 74 n. 52 – 6.403: 153 – 8.372 – 376: 74 – 75 – 9.361: 140 – 9.373 – 374: 140 – 11.439: 139 n. 27 – 11.498: 73 – 11.569: 117 – 118 – 11.619: 73 – 12.176: 73 – 12.69 – 70: 150 n. 19 – 12.95 – 97: 178 – 14.469 – 502: 105 n. 35 – 15.349: 73 – 17.224: 141 – 22.300: 81 – 22.376: 49 [Homer], Hymns – 2.6 – 9: 52 – 53 – 2.372: 82 n. 73 – 2.411 – 412: 82 – 83 n. 73 – 2.425 – 428: 52 – 53 – 4.449: 59 n. 15 – 4.512: 59 n. 15 – 5.73: 37 Horace Ars – 1 – 4: 200 – 201

Index of passages cited

– 1: 201 n. 59 – 3 – 4: 201 n. 59 – 10: 201 n. 59 – 15 – 16: 45 n. 74 – 301 – 302: 91 – 465 – 470: 200 – 476: 200 Epistles – 1.1.3: 29 n. 29 Epodes – 13.16: 146 n. 9 Odes – 1.1.33: 51 n. 91, 152 – 1.6.1 – 2: 199 n. 50 – 2.4.24: 29 n. 29 – 2.20.1 – 4: 198 – 199 – 2.20.9 – 12: 198 – 199 – 2.20.10 – 11: 201 Satires – 1.10.1: 151 n. 22 – 1.10.57 – 61: 151 n. 22 – 1.10.84 – 88: 151 – 2.5.4: 146 n. 9 – 2.7.49 – 50: 201 n. 59 – 2.7.55: 201 n. 59 – 2.7.59: 201 n. 59 – 2.7.62: 201 n. 59 – 2.7.64 – 65: 201 n. 59 Juvenal, Satires – 6.500: 183 Lucretius, De rerum natura – 1.15 – 16: 117 – 1.19: 33, 37, 54 – 1.36 – 40: 65 – 1.62 – 101: 137 – 1.64 – 65: 137 n. 23 – 1.72 – 74: 36, 71 – 72 – 1.73 – 74: 138 – 1.89 – 91: 130 n. 8 – 1.91: 131 n. 9 – 1.92: 133 – 1.98 – 101: 130 – 131 – 1.102 – 103: 131 – 1.471 – 477: 206 – 1.487 – 488: 206 n. 72

– 1.507 – 509: 206 – 1.718 – 719: 153 – 1.719: 5 – 1.922 – 930: 33, 35 – 1.924: 37 – 1.924 – 925: 54 – 1.925 – 926: 36 – 1.926 – 950: 4 n. 19 – 1.926 – 927: 37 – 2.7 – 14: 33 – 35 – 2.600 – 603: 72 – 2.600 – 601: 50 – 2.602 – 603: 75 – 2.628: 72 – 3.3 – 5: 117 – 3.526 – 539: 181, 193 – 3.528 – 529: 199 – 3.701: 183 – 3.726: 182 – 3.748 – 756: 207 – 3.755: 133 n. 14, 182 n. 15 – 3.868: 136 – 4.1 – 25: 4 n. 19 – 4.1037 – 1038: 54 – 4.1043: 54 – 4.1105: 201 n. 59 – 4.1117 – 1120: 91 – 4.1125: 91 n. 103 – 4.1177 – 1184: 53 – 54 – 5.7 – 12: 117 – 5.1198 – 1202: 132 – 5.1198: 131 n. 10 – 5.1209 – 1210: 132 – 5.1218: 131 n. 9 – 5.1379 – 1435: 59 – 5.1408: 59 n. 15 – 6.52: 131 n. 9 – 6.1239: 65 Manilius, Astronomica – 2.906: 96 n. 9 – 5.478: 50 n. 90 Martial, Epigrams – 4.31.7: 51 n. 91, 152

233

234

Index of passages cited

Moero (Powell) – fr. 1.5 – 8: 204 Moschus, Europa – 2 – 4: 100, 102, 166 – 7 – 8: 100 – 8: 166 – 41: 166 – 43 – 61: 165 – 58 – 61: 183 – 184 – 62: 166 – 65 – 71: 52 – 53 – 74 – 79: 163 – 165 – 74: 6, 164 – 77: 67 n. 35 – 93: 166 – 118 – 124: 149 – 129 – 130: 149 – 129: 166 – 131 – 132: 166 – 135 – 152: 149 – 135: 166 – 142 – 143: 166 – 146 – 148: 167 – 148: 163 – 152: 167 – 155 – 159: 166 – 161: 118 Nicander Alexipharmaca – 207 – 248: 86 – 213: 86 – 215 – 220: 86 – 87 – 242: 86 – 244 – 248: 88 – 89 Theriaca – 184: 88 – 266 – 270: 174 Nonnus, Dionysiaca – 20.21: 48 n. 82 – 47.392 – 393: 162 [Orpheus] fr. 34 (Kern): 77 – 78

Argonautica – 1 – 55: 38 – 47 – 51: 39 – 47: 41 Ovid Epistulae ex Ponto – 2.4.28: 53 n. 94 Fasti – 4.437: 53 n. 94 – 5.1 – 10: 50 – 51 – 5.5 – 6: 152 – 5.9: 152 – 5.53: 51 n. 91, 152 Heroides – 20.110: 130 – 21.81 – 82: 156 Metamorphoses – 1.15 – 17: 73 n. 50 – 1.20: 73 n. 50 – 1.266 – 271: 197 – 4.602 – 603: 118 n. 59 – 4.608: 118 n. 59 – 8.7 – 8: 58 – 8.14 – 18: 57 – 58 – 8.19 – 20: 58 – 8.20: 166 – 8.23: 166 – 8.49 – 50: 166 – 8.74 – 75: 82 – 8.81 – 82: 92 n. 1 – 8.82 – 84: 166 – 8.90: 83 – 8.98 – 100: 166 – 8.98: 206 n. 71 – 8.104 – 105: 166 – 8.108: 166 – 8.111 – 112: 166 – 8.118: 166 – 8.120: 166 – 8.122 – 123: 166 – 8.138: 166 – 8.141 – 142: 166 – 8.142 – 143: 150 n. 20 – 8.142: 127 – 8.314 – 315: 115 n. 57 – 11.339 – 341: 196

Index of passages cited

– 11.341: 195 n. 40 – 11.344 – 345: 196 – 14.59 – 63: 179 – 180 – 15.150 – 151: 36 – 15.160 – 162: 189 Tristia – 2.1.393 – 394: 18 n. 65 Parmenides (Diels – Kranz) – fr. 1.11 – 23: 37 – fr. 1.22: 37 n. 53 – fr. 1.27: 37 – fr. 6.3 – 7: 35 – 36 – fr. 7.3 – 5: 34 n. 41 – fr. 8.43: 75 n. 55 Parthenius (?) – SH 964.11: 64 Philodemus – AP 5.112: 26 – 28, 113 – AP 5.112.4 – 6: 32 – AP 5.120.1 – 2: 26 n. 22 – AP 11.41: 27, 29 n. 29 – AP 11.41.7 – 8: 29 Pindar, Pythian Odes – 4.233: 164 – 4.243: 142 – 143 n. 35 – 4.247 – 251: 142 – 144 – 4.247 – 248: 146

Plutarch, Life of Demetrius – 38.1 – 12: 84 Posidippus – SH 705 Propertius, Elegies – 3.5.32: 186 n. 18 – 4.4.1: 64, 82 n. 72 – 4.4.14: 108 n. 42 – 4.4.21: 108 n. 42 – 4.4.31 – 32: 108 n. 42 – 4.4.37: 107 – 4.4.38: 108 n. 42 – 4.4.63 – 64: 108 n. 42 – 4.4.64: 96 n. 9 – 4.4.66: 108 n. 42 – 4.4.69: 108 n. 42 – 4.4.79 – 80: 108 n. 42 – 4.4.85 – 86: 108 n. 42 – 4.4.88: 130 Sappho (Voigt) – fr. 31: 85 Scholia Ad Ap. Arg. 1.763: 43 Ad Ar. Ran. 125 – 126: 192 n. 33 Ad Hom. Il. 3.126 – 127: 45 Silius, Punica – 3.650: 118 n. 59

Plato Phaedo – 67c: 193 – 108a–b: 194 – 115d: 198 – 117e–118a: 192 – 193, 199 Phaedrus – 264c–e: 4 Symposium – 187d–e: 50 n. 88 – 197a–b: 171 – 207 – 209: 54

Suetonius / Donatus, Vita Vergilii – 46: 3

Plautus, Cistellaria – 298: 89

Theocritus, Idylls – 2.45 – 46: 159 n. 50

Sophocles, Trachiniae – 567: 89 – 573 – 574: 89 – 90 – 717: 89 – 767 – 771: 89 Statius, Thebaid – 9.401 – 403: 140

235

236

Index of passages cited

– 2.82: 6, 164 – 3.42: 6, 164 – 6.36 – 37: 140 – 6.39 – 40: 139 – 7.68 – 69: 189 n. 26 – 7.86: 189 n. 26 – 12.27 – 29: 60 – 61 – 13.24: 203 – 13.70 – 71: 81 n. 66 – 22.98 – 99: 140 – 22.213: 170 – 24.96: 142 – 24.98: 141 – 26.13: 79 – 26.16 – 17: 78 – 79 [Tibullus], Panegyricus Messallae – 3.7.11: 199 n. 49 – 3.7.17: 200 n. 53 – 3.7.179 – 180: 199 n. 50 – 3.7.196: 200 – 3.7.203: 200 – 3.7.204 – 211: 199 – 200 Valerius Flaccus, Argonautica – 3.422: 186 n. 18 – 4.49: 195 n. 40 – 5.531: 118 n. 59 – 5.534 – 537: 118 n. 59 Virgil Aeneid – 1.1: 55 – 1.8: 55 n. 6 – 1.456 – 493: 46 – 1.472 – 473: 108 n. 42 – 2.9: 96 n. 9 – 2.205 – 208: 187 – 3.73 – 74: 155 – 3.124 – 127: 156 – 3.426 – 427: 201 n. 59 – 4.81: 96 n. 9 – 4.91: 82

– 4.693 – 705: 194 – 4.701: 197 – 5.657 – 658: 195 n. 40 – 5.861: 196 – 197 – 6.378: 47 n. 81 – 6.406 – 407: 112 – 6.406: 5 – 6.658: 201 n. 59 – 8.36 – 37: 146 n. 9 – 8.80: 147 – 8.90 – 92: 147 – 8.219 – 220 – 8.291 – 293: 121 n. 68 – 8.578 – 583: 124 – 125 – 8.691 – 692: 156 n. 38 – 9.14 – 19: 195 – 197 – 9.731 – 733: 186 – 187 – 11.296 – 375: 118 n. 59 Eclogues – 2.50: 53 n. 94 – 6.1 – 12: 20 – 6.74 – 81: 159 n. 45 – 6.74 – 77: 49 – 8.41: 6, 164 – 10.69: 6, 169 Georgics – 1.338: 204 n. 64 – 1.406 – 409: 4, 195 – 2.477: 37 n. 53 – 3.90 – 91: 50 n. 90 – 3.152 – 153: 67 n. 35 – 4.97: 140 – 4.373: 186 n. 18 – 4.413: 133 n. 14, 182 n. 15, 207 n. 73 – 4.431: 5, 153 n. 30 – 4.443: 137 n. 24 – 4.511 – 515: 23 n. 16 – 4.559 – 566: 32 [Virgil], Catalepton – 5.5: 28 – 5.11 – 14: 28